This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Review ArticlePractical Aspects for the Integration of 5G Networks and IoTApplications in Smart Cities Environments
Daniel Minoli and Benedict Occhiogrosso
DVI Communications Inc New York NY USA
Correspondence should be addressed to Daniel Minoli danielminolidvicommcom
Received 1 April 2019 Accepted 26 May 2019 Published 5 August 2019
Guest Editor Xi Chen
Copyright copy 2019 DanielMinoli andBenedictOcchiogrossoThis is an open access article distributed under theCreativeCommonsAttribution License which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited
Both 5G cellular and IoT technologies are expected to seewidespread deployment in the next few years At the practical level 5Gwillsee initial deployments in urban areasThis is perhaps fortuitous from an IoT perspective sincemany ldquomainstreamrdquo applications ofIoT will support Smart Cities Smart Campuses and Smart Buildings Bandwidth demand for a number of Smart City applicationsis the main driver for enhancedmobile broadband (eMBB)-based 5G services in general and new-generation 5G IoT applicationsin particular In turn the use of the millimeter wave spectrum is required to enable 5G cellular technologies to support high datarates Millimeter wave solutions however impose a requirement for small cells Generally an implementer tries to use one or asmall handful of IoT technologies preferably and for managerial simplicity the implementer would want to use a cellular5G IoTtechnology for all nodes whether indoors or outdoors instead of a heterogenousmix of various IoT technologies that have evolvedover the yearsThis overviewpaper discusses a number of practical issues related to 5G-based IoT applications particularly in SmartCity environments including the need for small cells the transmission issues at millimeter wave frequencies building penetrationissues the need for Distributed Antenna Systems and the near term introduction of pre-5G IoT technologies such as NB-IoT andLTE-M these being possible proxies for the commercial deployment and acceptance of 5G IoT
1 Background
As the second decade of the 21st century comes to a closewe are witnessing an expansion of urban ecosystems aspopulations continue to sustain the transition from rural andsome suburban areas into large urban areas driven by eco-nomic opportunities demographic shifts and generationalpreferences Seventy percent of the human population isexpected to live in cities by the year 2050 and there alreadyare more than 400 cities with over one million inhabitants[1] Societal movement of people is a basic human existencedynamic and is one of the key mechanisms that drives thegrowth of cities Yet especially in the Western World citiesoften have aging infrastructure including roads bridgestunnels rail yards and power distribution plants It followsthat new technological solutions are needed to optimize theincreasingly-scarce infrastructure resources especially giventhe population growth and the limited financial resourcesavailable to most cities and municipalities When cities
deploy state-of-the-art Information and CommunicationTechnologies (ICT) on a large-scale including Internet ofThings (IoT) technologies they are referred to as beingldquoSmart Citiesrdquo
Livability infrastructure management asset manage-ment traffic transportation and mobility logistics electricpower and other utilities and physical security are thekey aspects of a cityrsquos operation IoT technologies offer theopportunity to improve resource management of many assetsrelated to city life and urban Quality of Life (QoL) includingIntelligent Transportation Systems (vehicular automationand traffic control) energy consumption the flow of goodssmart buildings spaceoccupancy management (indoors andoutdoors) pollution monitoring (for example from auto-mobile traffic factories incinerators crematoria) resourcemonitoring and sensing immersive services (including wear-ables and crowdsensing) physical security sustainability andthe greening of the environment Smart City IoT applica-tions cover indoor and outdoor applications they also span
HindawiWireless Communications and Mobile ComputingVolume 2019 Article ID 5710834 30 pageshttpsdoiorg10115520195710834
Figure 1 Illustrative example of Smart City resources that can benefit from IoT in general and 5G cellular in particular
stationary and mobile end-nodes and sensors There is anextensive body of literature on this topic some referencesof interest include but are certainly not limited to [2ndash12]Up to the present the IoT has been mostly utilized forsupporting a large population of relatively low-bandwidthsensing devices and where the sensing devices are typicallydeployed in stationary locations (eg electric meters build-ing management systems meteorological weather stations)However video-oriented applications that require streamsranging up to Ultra-High Definition resolution are emerging(eg surveillance physical security) In the evolving IoTenvironment the endpoint devices include environmentaland situational sensors vehicles wearables drones robotsand Virtual Reality gear In some applications IoT actuatorsare also utilized to control the physical ecosystem in responseto a sensed set of data or some analytical calculation ndash forexample changing the barriers and signs on a road to reversetraffic lanes during the day or changing the parameters ofa pump to control water or sewer flows Figure 1 depictsgraphically some of the common Smart City applications
While a large number of definitions and descriptionsof the IoT exist this is one definitionaldescriptive quote
from the authorrsquos previous work which we utilize hereldquoThe basic concept of the IoT is to enable objects of all kindsto have sensing actuating and communication capabilitiesso that locally-intrinsic or extrinsic data can be collectedprocessed transmitted concentrated and analyzed for eithercyber-physical goals at the collection point (or perhaps alongthe way) or for processenvironmentsystems analytics (of apredictive or historical nature) at a processing center often ldquointhe cloudrdquo Applications range from infrastructure and critical-infrastructure support (for example smart grid Smart Citysmart building and transportation) to end-user applicationssuch as e-health crowdsensing and further along to a mul-titude of other applications where only the imagination is thelimitrdquo [13ndash27] According to the Global System for MobileCommunications Association (GSMA) between 2018 and2025 the number of global IoT connections will triple to 25billion while global IoT revenue will quadruple to $11 trillion[28] others offer higher numbers (eg according to Statisticathere will be about 80 billion IoT devices worldwide in 2025[29])
5G (5th Generation) is the term for the next-generationcellularwireless service provider network that aims at
Wireless Communications and Mobile Computing 3
delivering higher data rates -- 100 times faster data speedsthan the current 4G Long Term Evolution (LTE) technology-- lower latency and highly-reliable connectivity In a senseit is an evolution of the previous generations of cellulartechnology
Smart Cities do not depend on any unique or specific IoTtechnology per se but include a panoply of IoT technologiessuch as mission-specific sensors appropriate networks andfunction-and-use-efficient analytics these often in the cloudWireless connectivity plays an important role in the utilityof this technology especially at the geographic scope of alarge or even medium-size city For practical reasons wirelessis also important in Smart Campus and Smart Buildingapplications Table 1 identifies a number of Smart Citychallenges and needs possible IoT-based solutions wirelessrequirements and the applicability of 5G solutions 5G IoTis licensed cellular IoT In this table ldquolow bandwidthrdquo equatesto 200 kbps or less ldquomedium bandwidthrdquo equates to 200kbps to 2 Mbps and ldquohigh bandwidthrdquo equates to morethan 2 Mbps Some IoT applications entail periodic ldquobatchrdquocommunication while other applications require real-timecommunication in the table ldquolow latencyrdquo means real-timeand ldquomedium latencyrdquomeans 1-to-5 seconds Table 2 providesa snapshot of key wireless technologies that are applicable tothe IoT environment A number of wireless technologies areavailable each with its specific applicability and functionalityThe direct use of traditional cellular services (eg 4GLTEnetworks) is not optimal for IoT applications both for costand nodal power-consumption reasons Furthermore theseservices are not ideal for a number of IoT applications wherea small amount of data is transmitted infrequently (egelectric gas or water meters for reading) Node density isalso an issue Cellular IoT solutions endeavor to addresslow-power low data rate requirements Several iterationsand alternatives solutions have emerged in recent years (egCat1Rel 8 Cat 0Rel 12 Cat-MRel 13 EC-GSM and NB-IoTRel 13) The 5G IoT system is the next evolutionary stepperhaps also affording some simplification and technologyhomogeneity
Figure 2 depicts the pre-5G and the 5G IoT connectivityecosystem which is further elaborated in the rest of thispaper The figure illustrates a typical case of Wi-Fi (in-building) aggregation of sensor data for a handoff to the cloudover a traditional router it illustrates the use of Low PowerWide Area Network (LPWAN) overlay technologies such asLoRa and Sigfox it shows the use of pre-5G IoT technologiesand then illustrates the use of 5G IoT in a native mode or ina more realistic Distributed Antenna System (DAS)-assistedmode
This review position and assessment paper provides anoverview of salient 5G features and then discusses somepractical design issues applicable to the IoT A lot of theimportant 5G IoT information is included in the figures andtables This paper is not intended to be a full 5G overview perse nor a discussion of IoT for both of which there are manyreferences (eg [30ndash34] for 5G and close to one hundredbooks on the IoT topic alone)
2 5G Concepts and Technology
5G cellular networks are now starting to be deployed aroundthe world as the underlying standards and the system-widetechnology become more mature (the term ldquoInternationalMobile Telecommunications-2020 [IMT-2020]rdquo is also used bythe standards bodies) Industry observers predict that societaldevelopments will lead to changes in the way communicationsystems are used and that these developments will in turnlead to a significant increase inmobile andwireless traffic vol-ume such traffic volume is expected to increase a thousand-fold over the next decade Observations such as this one arecommon in the literature positioning the technology ldquoUnlikeprevious generations of mobile networks the fifth generation(5G) technology is expected to fundamentally transform the rolethat telecommunications technology plays in the societyrdquo [34]
The 5G system expands the 4G environment by addingNew Radio (NR) capabilities but doing so in such a mannerthat LTE and NR can evolve in complementary ways As itmight be envisioned a 5G system entails devices connectedto a 5G access network which in turn is connected to a5G core network The 5G access network may include 3GPP(3rd Generation Partnership Project) radio base stationsandor a non-3GPP access network The 5G core networkoffers major improvements compared with a 4G system inthe area of network slicing and service-based architectures(SBAs) in particular the core is designed to support cloudimplementation and the IoT 5G systems subsume important4G system concepts such as the energy saving capabilitiesof narrowband IoT (NB-IoT) radios secure low latencysmall data transmission for low-power devices -- low latencyis a requirement for making autonomous vehicles safe --and devices using energy-preserving dormant states whenpossible Network slicing allows service providers to deliverldquoNetwork as a Service (NaaS)rdquo to largeinstitutional usersaffording them the flexibility to manage their own servicesand devices on the 5G providerrsquos network
Applications driving wireless traffic include but arenot limited to on-demand mobile information and high-resolution entertainment augmented reality virtual realityand immersive services e-health and ubiquitous IoT roll-outsWhile 5G technology could still take several distinct ser-vice directions it appears at this juncture that the view favor-ing a super-fast mobile network where densely-clusteredsmall cells provide contiguous urban coverage to mobile aswell as stationary users is the approach envisioned by thestandards development bodies and by the implementers Itshould be noted that in the US upwards of 55 percent ofresidential users now utilize cellular-services-only at home inplace of a landline and about 30 percent of residential usersutilize both with the trend favoring an eventual transitionto the former Therefore the evolving 5G systems will haveto properly support this growing segment of the market Agoal of 5G networks is to be five times as fast as comparedto the highest current speed of existing 4G networks withdownload speeds as high as 5 Gbps ndash 4G offering only up to amaximum of 1 Gbps Deployment of 5G networks started in2018 in some advanced countries although further develop-ments on fundamentals will continue naturally the current
4 Wireless Communications and Mobile Computing
Table1Ke
yUrban
Challenges
andIoT-supp
ortedSolutio
ns
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Infrastructureandrealestate
managem
ent
Requ
irementmon
itorstatusa
ndoccupancyo
fspacesbu
ildingsroads
bridgestunn
elsrailroadcrossin
gsand
streetsignals
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)toprovider
eal-tim
eand
histo
ricaltre
ndingdataallowingcity
agencies
toprovidee
nhancedvisib
ility
into
thep
erform
ance
ofresources
facilitatingenvironm
entaland
safety
sensingsm
artp
arking
andsm
artp
arking
meterssm
artelectric
metersandsm
art
build
ingfunctio
nality
YY
High
Low
Low
Medium
Livability
Requ
irementQualityof
Lifeexp
editiou
saccessto
servicesefficienttranspo
rtation
lowdelayssafety
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)tofacilitates
martm
ulti-mod
altransportatio
ninform
ation-ric
henvironm
ents
locatio
n-basedservices
real-timec
onnectivity
tohealth-m
onito
ringresources(eg
air
quality
)
YY
High
Medium
Medium
Medium
Logistics
Requ
irementsupp
lyingcitydw
ellers
with
fresh
food
sup
pliesgood
sand
otherm
aterials
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)toenablethes
tream
liningof
warehou
singtransportatio
nand
distr
ibutionof
good
sTraffi
cmanagem
entisa
faceto
fsuchlogistical
supp
ort
YY
High
MediumM
edium
High
Physicalsecurity
Requ
irementsecurityinstr
eets
parks
statio
nstun
nels
bridgestrainsbuses
ferries
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
esandgu
nsho
tdetectio
nsyste
ms)to
supp
ortIP-basedsurveillancev
ideo
license
plater
eading
gun
-sho
tdetectio
nbio-hazard
andradiological
contam
inationmon
itorin
gface
recogn
ition
and
crow
dmon
itorin
gand
control
Perhaps
YHigh
High
Low
High
Wireless Communications and Mobile Computing 5
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Powe
rand
otherc
ity-sup
portingutilitie
s
Requ
irementreliablefl
owof
electric
energygasand
wateroptim
ized
waste-m
anagem
entand
sewe
rsafe
storage
ofgasolin
e
SmartG
ridsolutio
nsandsensor-rich
utilityinfrastructure
NY
High
Low
Medium
High
Traffi
ctransportatio
nandmob
ility
Requ
irementop
timized
traffi
cflow
low
congestio
nlowlatencya
ndhigh
expediencylow
noise
minim
alwasteof
fuelandCO2em
issionssafety
Netwo
rked
sensorstosupp
orttrafficfl
ow
driverlessvehiclesinclu
ding
driverless
bustransit
andmulti-mod
altransportatio
nsyste
msFo
rdriv
erless
vehicles
sensorsw
illallow
high
-resolutionmapping
telem
etry
data
traffi
cand
hazard
avoidancem
echanism
s
NY
High
Medium-to
-High
Low
High
Electricandotheru
tility
manho
lemon
itorin
g
Requ
irementElectricpo
werm
anho
les
requ
iremon
itorin
gto
avoidandor
preventd
angerous
situatio
ns
Cost-e
ffectivea
ndreliables
ensorsare
neededTechn
olog
ybeing
investigatedby
Con
Ediso
nin
New
York
city
NY
High
Low
Medium
High
Pollu
tionmon
itorin
g
Requ
irementmon
itore
missionof
dioxinsvapo
rized
mercury
nano
particlesradiationfro
mfactories
incineratorsurban
crem
atoriaespecially
iftheses
ources
arec
lose
totraintracks
orotherw
ind-turbulence
elem
ents(eg
canyon
s)
Netwo
rked
sensorsthrou
ghou
ttow
n(or
with
in10
kmof
apoint
source)to
mon
itortoxichealth
-impactingem
ission
from
pointsou
rces
inclu
ding
factories
generatio
nplants(if
any)
andcrem
atoria
(ifany)
[35ndash46
]
NY
High
MediumM
edium
High
6 Wireless Communications and Mobile Computing
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Environm
entalM
onito
ring
Requ
irements
mon
itoro
utdo
ortemperaturehum
idity
andother
environm
entalgases
Sensorstothatcanbe
placed
ineasy-to
-deploylocatio
nsegatop
existingSm
artC
itylig
htpo
lesto
continuo
uslymon
itortem
perature
humidity
andothere
nviro
nmentalgases
NY
High
Low
MediumM
edium
Floo
dAb
atem
ent
Requ
irementFloo
dandsto
rmdrainage
control
Distrib
uted
ruggedized
sensorsto
mon
itorF
lood
andsto
rmdrainage
toprovidee
arlywarning
andfaultd
etectio
nN
YHigh
Low
Medium
High
SmartC
ityLigh
ting
Requ
irementCon
versionto
LED
lightingandensuingcontrolviaIoTfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
Citie
sspend
largea
mou
ntso
fmon
eyyearlyforstre
etlig
hting(usually1000
streetlightsp
er10000
inhabitantsand
$125
pery
earp
erlig
htfor4
662ho
urso
fusagey
early
andsyste
mam
ortization)
LEDlig
htingrequ
ires13rd
thea
mou
ntof
powe
rfor
thes
amea
mou
ntof
luminance
Paybackforc
onversionisno
warou
nd5-6
yearsSensorsa
reneeded
for
IoT-directed
light
managem
entfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
NY
High
Medium
Medium
Medium
Wireless Communications and Mobile Computing 7
Table2Ke
yWire
lessTechno
logies
applicableto
IoT
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
5GYesperhapsw
ithDistrib
uted
Antenna
Syste
ms(DASs)
Yesabou
t10-15
miles
(i)Evolving
not
yetw
idely
deployed
(ii)S
everalband
slowlatencyhigh
sensor
density
(iii)Cellularn
etwo
rkarchitecture
(iv)L
icensedspectrum
001M
bpsinsomeimplem
entatio
nsbattery
sim10years
(v)B
roadband
features
availablefor
surveillancemultim
edia
(vi)Cost-e
ffective
(vii)
Expected
tobe
availablew
orldwide
(viii)B
uildingpenetrationmay
need
Distrib
uted
Antenna
Syste
ms
(DASs)
NB-IoT
(Narrowband
IoT)
Yes
Yesup
toabou
t20m
iles
(i)Severalbandslicensedspectrum
(ii)L
TE-based
(iii)01-0
2Mbp
sdatar
atesbatterysim10
+years
(iv)L
owcost
lowmod
emcomplexitylow
powe
renergy
saving
mechanism
s(high
batte
rylife)
(v)D
oesn
otrequ
ireag
atew
aysensord
ataissentd
irectlyto
the
destinatio
nserver
(other
IoTsyste
mstypicallyhave
gatewaysthat
aggregates
ensord
atawhich
then
commun
icatew
iththed
estin
ation
server)
(vi)Re
ason
ablebu
ildingpenetration(im
proved
indo
orcoverage)
(vii)
Largen
umbero
flow
throug
hput
devices(up
to15000
0devices
perc
ell)
8 Wireless Communications and Mobile Computing
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
LTE-M
(Lon
g-Term
EvolutionMachine
Type
Com
mun
ications)
Rel13(C
atM1C
atM)
Yes
Yesabou
t10-20
miles
(i)Cellularn
etwo
rkarchitectureLT
Ecompatib
leeasyto
deployn
ewcellu
lara
ntennasn
otrequ
ired
(ii)U
ses4
G-LTE
band
sbelow
1GHzlicensedspectrum
(iii)Con
sidered
thes
econ
dgeneratio
nof
LTEchipsa
imed
atIoT
applications
(iv)C
apsm
axim
umsyste
mband
width
at14
MHz(
asop
posedto
Cat-0rsquos20
MHz)thu
sisc
ost-e
ffectivefor
LowPo
werW
ideA
rea
Netwo
rk(LPW
AN)app
lications
such
assm
artm
eteringwhereon
lysm
allamou
ntof
datatransfe
risrequired
(v)1
Mbp
suploaddo
wnload
batte
rysim10
years
(vi)Re
lativ
elylowcomplexity
andlowpo
werm
odem
(vii)
Can
beused
fortrackingmovingob
jects(Lo
catio
nservices
provided
throug
hcelltowe
rmechanism
s)
LoRa
Yes
Yes(6-15
milesw
ithLO
S)
(i)Ba
ndbelow1G
Hz
(ii)IoT
-focusedfro
mtheg
et-go
(iii)Prop
rietary
(iv)L
owpo
wer
Sigfox
Somew
hatlim
ited
Yes(30
milesinrural
environm
ents
1-6miles
incityenvironm
ents)
(i)Ba
ndbelow1G
Hz
(ii)N
arrowband
(iii)Lo
wpo
wer
(iv)S
tartop
olog
y
Wireless Communications and Mobile Computing 9
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
Wi-F
iYes300feet
Tosomed
egreerequ
ires
inter-spot
conn
ectiv
itybackbo
ne(w
iredor
wire
less)(eg
80211ah
dista
ncer
ange
upto
abou
t12
mile)
(i)Severalbands
(ii)In2018
theF
CCallowe
dthee
xpansio
nof
the6
GHzb
andto
next-generationWi-F
idevices
with
12GHzo
fadd
ition
alspectrum
spanning
5925to
7125
GHz(
currentW
i-Fin
etwo
rkso
perateat24
GHza
nd5GHzw
ithafew
vend
orso
fferin
g60
GHzldquo
WiGigrdquothis
having
arange
of30
feetndashIEEE
80211a
dandIEEE
80211a
y)(iii)Highadop
tion
most(bu
tnot
all)indo
orIoTutilize
Wi-F
igood
functio
nality
(iv)F
reeldquo
airtim
erdquo(v)S
ubjectto
interfe
rencemalicious
orno
n-malicious
interfe
rence
(egtoo
manyho
tspo
ts)couldim
pairthes
ensorfrom
send
ingdata
either
onafi
ne-grain
orcoarse-grain
basis
Bluetooth
Yes30
feet
No(orfor
Person
alArea
only)
(i)Lo
wband
width
(2Mbp
s)(ii)U
sedin
medicaldevicesa
ndindu
strialsensorsLo
wpo
wergood
forw
earables
(iii)Usablefor
Realtim
elocationsyste
msw
ithmedium
accuracy
Zigbee
Yes(30-300
feet)
No(orfor
Person
alArea
only)
(i)Lo
wdatarate
(ii)Ind
ustrialand
someh
omea
pplications
(egho
mee
nergy
mon
itorin
gwire
lesslig
htsw
itches)
(iii)Lo
wtransm
itpo
werLo
wbatte
ryconsum
ption
NoteAfewotherlegacyIoTwire
lesstechno
logies
exist
(egCat0Cat1EC
-GSM
Weightless)b
utaren
otinclu
dedin
thistable
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Figure 1 Illustrative example of Smart City resources that can benefit from IoT in general and 5G cellular in particular
stationary and mobile end-nodes and sensors There is anextensive body of literature on this topic some referencesof interest include but are certainly not limited to [2ndash12]Up to the present the IoT has been mostly utilized forsupporting a large population of relatively low-bandwidthsensing devices and where the sensing devices are typicallydeployed in stationary locations (eg electric meters build-ing management systems meteorological weather stations)However video-oriented applications that require streamsranging up to Ultra-High Definition resolution are emerging(eg surveillance physical security) In the evolving IoTenvironment the endpoint devices include environmentaland situational sensors vehicles wearables drones robotsand Virtual Reality gear In some applications IoT actuatorsare also utilized to control the physical ecosystem in responseto a sensed set of data or some analytical calculation ndash forexample changing the barriers and signs on a road to reversetraffic lanes during the day or changing the parameters ofa pump to control water or sewer flows Figure 1 depictsgraphically some of the common Smart City applications
While a large number of definitions and descriptionsof the IoT exist this is one definitionaldescriptive quote
from the authorrsquos previous work which we utilize hereldquoThe basic concept of the IoT is to enable objects of all kindsto have sensing actuating and communication capabilitiesso that locally-intrinsic or extrinsic data can be collectedprocessed transmitted concentrated and analyzed for eithercyber-physical goals at the collection point (or perhaps alongthe way) or for processenvironmentsystems analytics (of apredictive or historical nature) at a processing center often ldquointhe cloudrdquo Applications range from infrastructure and critical-infrastructure support (for example smart grid Smart Citysmart building and transportation) to end-user applicationssuch as e-health crowdsensing and further along to a mul-titude of other applications where only the imagination is thelimitrdquo [13ndash27] According to the Global System for MobileCommunications Association (GSMA) between 2018 and2025 the number of global IoT connections will triple to 25billion while global IoT revenue will quadruple to $11 trillion[28] others offer higher numbers (eg according to Statisticathere will be about 80 billion IoT devices worldwide in 2025[29])
5G (5th Generation) is the term for the next-generationcellularwireless service provider network that aims at
Wireless Communications and Mobile Computing 3
delivering higher data rates -- 100 times faster data speedsthan the current 4G Long Term Evolution (LTE) technology-- lower latency and highly-reliable connectivity In a senseit is an evolution of the previous generations of cellulartechnology
Smart Cities do not depend on any unique or specific IoTtechnology per se but include a panoply of IoT technologiessuch as mission-specific sensors appropriate networks andfunction-and-use-efficient analytics these often in the cloudWireless connectivity plays an important role in the utilityof this technology especially at the geographic scope of alarge or even medium-size city For practical reasons wirelessis also important in Smart Campus and Smart Buildingapplications Table 1 identifies a number of Smart Citychallenges and needs possible IoT-based solutions wirelessrequirements and the applicability of 5G solutions 5G IoTis licensed cellular IoT In this table ldquolow bandwidthrdquo equatesto 200 kbps or less ldquomedium bandwidthrdquo equates to 200kbps to 2 Mbps and ldquohigh bandwidthrdquo equates to morethan 2 Mbps Some IoT applications entail periodic ldquobatchrdquocommunication while other applications require real-timecommunication in the table ldquolow latencyrdquo means real-timeand ldquomedium latencyrdquomeans 1-to-5 seconds Table 2 providesa snapshot of key wireless technologies that are applicable tothe IoT environment A number of wireless technologies areavailable each with its specific applicability and functionalityThe direct use of traditional cellular services (eg 4GLTEnetworks) is not optimal for IoT applications both for costand nodal power-consumption reasons Furthermore theseservices are not ideal for a number of IoT applications wherea small amount of data is transmitted infrequently (egelectric gas or water meters for reading) Node density isalso an issue Cellular IoT solutions endeavor to addresslow-power low data rate requirements Several iterationsand alternatives solutions have emerged in recent years (egCat1Rel 8 Cat 0Rel 12 Cat-MRel 13 EC-GSM and NB-IoTRel 13) The 5G IoT system is the next evolutionary stepperhaps also affording some simplification and technologyhomogeneity
Figure 2 depicts the pre-5G and the 5G IoT connectivityecosystem which is further elaborated in the rest of thispaper The figure illustrates a typical case of Wi-Fi (in-building) aggregation of sensor data for a handoff to the cloudover a traditional router it illustrates the use of Low PowerWide Area Network (LPWAN) overlay technologies such asLoRa and Sigfox it shows the use of pre-5G IoT technologiesand then illustrates the use of 5G IoT in a native mode or ina more realistic Distributed Antenna System (DAS)-assistedmode
This review position and assessment paper provides anoverview of salient 5G features and then discusses somepractical design issues applicable to the IoT A lot of theimportant 5G IoT information is included in the figures andtables This paper is not intended to be a full 5G overview perse nor a discussion of IoT for both of which there are manyreferences (eg [30ndash34] for 5G and close to one hundredbooks on the IoT topic alone)
2 5G Concepts and Technology
5G cellular networks are now starting to be deployed aroundthe world as the underlying standards and the system-widetechnology become more mature (the term ldquoInternationalMobile Telecommunications-2020 [IMT-2020]rdquo is also used bythe standards bodies) Industry observers predict that societaldevelopments will lead to changes in the way communicationsystems are used and that these developments will in turnlead to a significant increase inmobile andwireless traffic vol-ume such traffic volume is expected to increase a thousand-fold over the next decade Observations such as this one arecommon in the literature positioning the technology ldquoUnlikeprevious generations of mobile networks the fifth generation(5G) technology is expected to fundamentally transform the rolethat telecommunications technology plays in the societyrdquo [34]
The 5G system expands the 4G environment by addingNew Radio (NR) capabilities but doing so in such a mannerthat LTE and NR can evolve in complementary ways As itmight be envisioned a 5G system entails devices connectedto a 5G access network which in turn is connected to a5G core network The 5G access network may include 3GPP(3rd Generation Partnership Project) radio base stationsandor a non-3GPP access network The 5G core networkoffers major improvements compared with a 4G system inthe area of network slicing and service-based architectures(SBAs) in particular the core is designed to support cloudimplementation and the IoT 5G systems subsume important4G system concepts such as the energy saving capabilitiesof narrowband IoT (NB-IoT) radios secure low latencysmall data transmission for low-power devices -- low latencyis a requirement for making autonomous vehicles safe --and devices using energy-preserving dormant states whenpossible Network slicing allows service providers to deliverldquoNetwork as a Service (NaaS)rdquo to largeinstitutional usersaffording them the flexibility to manage their own servicesand devices on the 5G providerrsquos network
Applications driving wireless traffic include but arenot limited to on-demand mobile information and high-resolution entertainment augmented reality virtual realityand immersive services e-health and ubiquitous IoT roll-outsWhile 5G technology could still take several distinct ser-vice directions it appears at this juncture that the view favor-ing a super-fast mobile network where densely-clusteredsmall cells provide contiguous urban coverage to mobile aswell as stationary users is the approach envisioned by thestandards development bodies and by the implementers Itshould be noted that in the US upwards of 55 percent ofresidential users now utilize cellular-services-only at home inplace of a landline and about 30 percent of residential usersutilize both with the trend favoring an eventual transitionto the former Therefore the evolving 5G systems will haveto properly support this growing segment of the market Agoal of 5G networks is to be five times as fast as comparedto the highest current speed of existing 4G networks withdownload speeds as high as 5 Gbps ndash 4G offering only up to amaximum of 1 Gbps Deployment of 5G networks started in2018 in some advanced countries although further develop-ments on fundamentals will continue naturally the current
4 Wireless Communications and Mobile Computing
Table1Ke
yUrban
Challenges
andIoT-supp
ortedSolutio
ns
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Infrastructureandrealestate
managem
ent
Requ
irementmon
itorstatusa
ndoccupancyo
fspacesbu
ildingsroads
bridgestunn
elsrailroadcrossin
gsand
streetsignals
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)toprovider
eal-tim
eand
histo
ricaltre
ndingdataallowingcity
agencies
toprovidee
nhancedvisib
ility
into
thep
erform
ance
ofresources
facilitatingenvironm
entaland
safety
sensingsm
artp
arking
andsm
artp
arking
meterssm
artelectric
metersandsm
art
build
ingfunctio
nality
YY
High
Low
Low
Medium
Livability
Requ
irementQualityof
Lifeexp
editiou
saccessto
servicesefficienttranspo
rtation
lowdelayssafety
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)tofacilitates
martm
ulti-mod
altransportatio
ninform
ation-ric
henvironm
ents
locatio
n-basedservices
real-timec
onnectivity
tohealth-m
onito
ringresources(eg
air
quality
)
YY
High
Medium
Medium
Medium
Logistics
Requ
irementsupp
lyingcitydw
ellers
with
fresh
food
sup
pliesgood
sand
otherm
aterials
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)toenablethes
tream
liningof
warehou
singtransportatio
nand
distr
ibutionof
good
sTraffi
cmanagem
entisa
faceto
fsuchlogistical
supp
ort
YY
High
MediumM
edium
High
Physicalsecurity
Requ
irementsecurityinstr
eets
parks
statio
nstun
nels
bridgestrainsbuses
ferries
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
esandgu
nsho
tdetectio
nsyste
ms)to
supp
ortIP-basedsurveillancev
ideo
license
plater
eading
gun
-sho
tdetectio
nbio-hazard
andradiological
contam
inationmon
itorin
gface
recogn
ition
and
crow
dmon
itorin
gand
control
Perhaps
YHigh
High
Low
High
Wireless Communications and Mobile Computing 5
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Powe
rand
otherc
ity-sup
portingutilitie
s
Requ
irementreliablefl
owof
electric
energygasand
wateroptim
ized
waste-m
anagem
entand
sewe
rsafe
storage
ofgasolin
e
SmartG
ridsolutio
nsandsensor-rich
utilityinfrastructure
NY
High
Low
Medium
High
Traffi
ctransportatio
nandmob
ility
Requ
irementop
timized
traffi
cflow
low
congestio
nlowlatencya
ndhigh
expediencylow
noise
minim
alwasteof
fuelandCO2em
issionssafety
Netwo
rked
sensorstosupp
orttrafficfl
ow
driverlessvehiclesinclu
ding
driverless
bustransit
andmulti-mod
altransportatio
nsyste
msFo
rdriv
erless
vehicles
sensorsw
illallow
high
-resolutionmapping
telem
etry
data
traffi
cand
hazard
avoidancem
echanism
s
NY
High
Medium-to
-High
Low
High
Electricandotheru
tility
manho
lemon
itorin
g
Requ
irementElectricpo
werm
anho
les
requ
iremon
itorin
gto
avoidandor
preventd
angerous
situatio
ns
Cost-e
ffectivea
ndreliables
ensorsare
neededTechn
olog
ybeing
investigatedby
Con
Ediso
nin
New
York
city
NY
High
Low
Medium
High
Pollu
tionmon
itorin
g
Requ
irementmon
itore
missionof
dioxinsvapo
rized
mercury
nano
particlesradiationfro
mfactories
incineratorsurban
crem
atoriaespecially
iftheses
ources
arec
lose
totraintracks
orotherw
ind-turbulence
elem
ents(eg
canyon
s)
Netwo
rked
sensorsthrou
ghou
ttow
n(or
with
in10
kmof
apoint
source)to
mon
itortoxichealth
-impactingem
ission
from
pointsou
rces
inclu
ding
factories
generatio
nplants(if
any)
andcrem
atoria
(ifany)
[35ndash46
]
NY
High
MediumM
edium
High
6 Wireless Communications and Mobile Computing
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Environm
entalM
onito
ring
Requ
irements
mon
itoro
utdo
ortemperaturehum
idity
andother
environm
entalgases
Sensorstothatcanbe
placed
ineasy-to
-deploylocatio
nsegatop
existingSm
artC
itylig
htpo
lesto
continuo
uslymon
itortem
perature
humidity
andothere
nviro
nmentalgases
NY
High
Low
MediumM
edium
Floo
dAb
atem
ent
Requ
irementFloo
dandsto
rmdrainage
control
Distrib
uted
ruggedized
sensorsto
mon
itorF
lood
andsto
rmdrainage
toprovidee
arlywarning
andfaultd
etectio
nN
YHigh
Low
Medium
High
SmartC
ityLigh
ting
Requ
irementCon
versionto
LED
lightingandensuingcontrolviaIoTfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
Citie
sspend
largea
mou
ntso
fmon
eyyearlyforstre
etlig
hting(usually1000
streetlightsp
er10000
inhabitantsand
$125
pery
earp
erlig
htfor4
662ho
urso
fusagey
early
andsyste
mam
ortization)
LEDlig
htingrequ
ires13rd
thea
mou
ntof
powe
rfor
thes
amea
mou
ntof
luminance
Paybackforc
onversionisno
warou
nd5-6
yearsSensorsa
reneeded
for
IoT-directed
light
managem
entfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
NY
High
Medium
Medium
Medium
Wireless Communications and Mobile Computing 7
Table2Ke
yWire
lessTechno
logies
applicableto
IoT
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
5GYesperhapsw
ithDistrib
uted
Antenna
Syste
ms(DASs)
Yesabou
t10-15
miles
(i)Evolving
not
yetw
idely
deployed
(ii)S
everalband
slowlatencyhigh
sensor
density
(iii)Cellularn
etwo
rkarchitecture
(iv)L
icensedspectrum
001M
bpsinsomeimplem
entatio
nsbattery
sim10years
(v)B
roadband
features
availablefor
surveillancemultim
edia
(vi)Cost-e
ffective
(vii)
Expected
tobe
availablew
orldwide
(viii)B
uildingpenetrationmay
need
Distrib
uted
Antenna
Syste
ms
(DASs)
NB-IoT
(Narrowband
IoT)
Yes
Yesup
toabou
t20m
iles
(i)Severalbandslicensedspectrum
(ii)L
TE-based
(iii)01-0
2Mbp
sdatar
atesbatterysim10
+years
(iv)L
owcost
lowmod
emcomplexitylow
powe
renergy
saving
mechanism
s(high
batte
rylife)
(v)D
oesn
otrequ
ireag
atew
aysensord
ataissentd
irectlyto
the
destinatio
nserver
(other
IoTsyste
mstypicallyhave
gatewaysthat
aggregates
ensord
atawhich
then
commun
icatew
iththed
estin
ation
server)
(vi)Re
ason
ablebu
ildingpenetration(im
proved
indo
orcoverage)
(vii)
Largen
umbero
flow
throug
hput
devices(up
to15000
0devices
perc
ell)
8 Wireless Communications and Mobile Computing
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
LTE-M
(Lon
g-Term
EvolutionMachine
Type
Com
mun
ications)
Rel13(C
atM1C
atM)
Yes
Yesabou
t10-20
miles
(i)Cellularn
etwo
rkarchitectureLT
Ecompatib
leeasyto
deployn
ewcellu
lara
ntennasn
otrequ
ired
(ii)U
ses4
G-LTE
band
sbelow
1GHzlicensedspectrum
(iii)Con
sidered
thes
econ
dgeneratio
nof
LTEchipsa
imed
atIoT
applications
(iv)C
apsm
axim
umsyste
mband
width
at14
MHz(
asop
posedto
Cat-0rsquos20
MHz)thu
sisc
ost-e
ffectivefor
LowPo
werW
ideA
rea
Netwo
rk(LPW
AN)app
lications
such
assm
artm
eteringwhereon
lysm
allamou
ntof
datatransfe
risrequired
(v)1
Mbp
suploaddo
wnload
batte
rysim10
years
(vi)Re
lativ
elylowcomplexity
andlowpo
werm
odem
(vii)
Can
beused
fortrackingmovingob
jects(Lo
catio
nservices
provided
throug
hcelltowe
rmechanism
s)
LoRa
Yes
Yes(6-15
milesw
ithLO
S)
(i)Ba
ndbelow1G
Hz
(ii)IoT
-focusedfro
mtheg
et-go
(iii)Prop
rietary
(iv)L
owpo
wer
Sigfox
Somew
hatlim
ited
Yes(30
milesinrural
environm
ents
1-6miles
incityenvironm
ents)
(i)Ba
ndbelow1G
Hz
(ii)N
arrowband
(iii)Lo
wpo
wer
(iv)S
tartop
olog
y
Wireless Communications and Mobile Computing 9
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
Wi-F
iYes300feet
Tosomed
egreerequ
ires
inter-spot
conn
ectiv
itybackbo
ne(w
iredor
wire
less)(eg
80211ah
dista
ncer
ange
upto
abou
t12
mile)
(i)Severalbands
(ii)In2018
theF
CCallowe
dthee
xpansio
nof
the6
GHzb
andto
next-generationWi-F
idevices
with
12GHzo
fadd
ition
alspectrum
spanning
5925to
7125
GHz(
currentW
i-Fin
etwo
rkso
perateat24
GHza
nd5GHzw
ithafew
vend
orso
fferin
g60
GHzldquo
WiGigrdquothis
having
arange
of30
feetndashIEEE
80211a
dandIEEE
80211a
y)(iii)Highadop
tion
most(bu
tnot
all)indo
orIoTutilize
Wi-F
igood
functio
nality
(iv)F
reeldquo
airtim
erdquo(v)S
ubjectto
interfe
rencemalicious
orno
n-malicious
interfe
rence
(egtoo
manyho
tspo
ts)couldim
pairthes
ensorfrom
send
ingdata
either
onafi
ne-grain
orcoarse-grain
basis
Bluetooth
Yes30
feet
No(orfor
Person
alArea
only)
(i)Lo
wband
width
(2Mbp
s)(ii)U
sedin
medicaldevicesa
ndindu
strialsensorsLo
wpo
wergood
forw
earables
(iii)Usablefor
Realtim
elocationsyste
msw
ithmedium
accuracy
Zigbee
Yes(30-300
feet)
No(orfor
Person
alArea
only)
(i)Lo
wdatarate
(ii)Ind
ustrialand
someh
omea
pplications
(egho
mee
nergy
mon
itorin
gwire
lesslig
htsw
itches)
(iii)Lo
wtransm
itpo
werLo
wbatte
ryconsum
ption
NoteAfewotherlegacyIoTwire
lesstechno
logies
exist
(egCat0Cat1EC
-GSM
Weightless)b
utaren
otinclu
dedin
thistable
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 3
delivering higher data rates -- 100 times faster data speedsthan the current 4G Long Term Evolution (LTE) technology-- lower latency and highly-reliable connectivity In a senseit is an evolution of the previous generations of cellulartechnology
Smart Cities do not depend on any unique or specific IoTtechnology per se but include a panoply of IoT technologiessuch as mission-specific sensors appropriate networks andfunction-and-use-efficient analytics these often in the cloudWireless connectivity plays an important role in the utilityof this technology especially at the geographic scope of alarge or even medium-size city For practical reasons wirelessis also important in Smart Campus and Smart Buildingapplications Table 1 identifies a number of Smart Citychallenges and needs possible IoT-based solutions wirelessrequirements and the applicability of 5G solutions 5G IoTis licensed cellular IoT In this table ldquolow bandwidthrdquo equatesto 200 kbps or less ldquomedium bandwidthrdquo equates to 200kbps to 2 Mbps and ldquohigh bandwidthrdquo equates to morethan 2 Mbps Some IoT applications entail periodic ldquobatchrdquocommunication while other applications require real-timecommunication in the table ldquolow latencyrdquo means real-timeand ldquomedium latencyrdquomeans 1-to-5 seconds Table 2 providesa snapshot of key wireless technologies that are applicable tothe IoT environment A number of wireless technologies areavailable each with its specific applicability and functionalityThe direct use of traditional cellular services (eg 4GLTEnetworks) is not optimal for IoT applications both for costand nodal power-consumption reasons Furthermore theseservices are not ideal for a number of IoT applications wherea small amount of data is transmitted infrequently (egelectric gas or water meters for reading) Node density isalso an issue Cellular IoT solutions endeavor to addresslow-power low data rate requirements Several iterationsand alternatives solutions have emerged in recent years (egCat1Rel 8 Cat 0Rel 12 Cat-MRel 13 EC-GSM and NB-IoTRel 13) The 5G IoT system is the next evolutionary stepperhaps also affording some simplification and technologyhomogeneity
Figure 2 depicts the pre-5G and the 5G IoT connectivityecosystem which is further elaborated in the rest of thispaper The figure illustrates a typical case of Wi-Fi (in-building) aggregation of sensor data for a handoff to the cloudover a traditional router it illustrates the use of Low PowerWide Area Network (LPWAN) overlay technologies such asLoRa and Sigfox it shows the use of pre-5G IoT technologiesand then illustrates the use of 5G IoT in a native mode or ina more realistic Distributed Antenna System (DAS)-assistedmode
This review position and assessment paper provides anoverview of salient 5G features and then discusses somepractical design issues applicable to the IoT A lot of theimportant 5G IoT information is included in the figures andtables This paper is not intended to be a full 5G overview perse nor a discussion of IoT for both of which there are manyreferences (eg [30ndash34] for 5G and close to one hundredbooks on the IoT topic alone)
2 5G Concepts and Technology
5G cellular networks are now starting to be deployed aroundthe world as the underlying standards and the system-widetechnology become more mature (the term ldquoInternationalMobile Telecommunications-2020 [IMT-2020]rdquo is also used bythe standards bodies) Industry observers predict that societaldevelopments will lead to changes in the way communicationsystems are used and that these developments will in turnlead to a significant increase inmobile andwireless traffic vol-ume such traffic volume is expected to increase a thousand-fold over the next decade Observations such as this one arecommon in the literature positioning the technology ldquoUnlikeprevious generations of mobile networks the fifth generation(5G) technology is expected to fundamentally transform the rolethat telecommunications technology plays in the societyrdquo [34]
The 5G system expands the 4G environment by addingNew Radio (NR) capabilities but doing so in such a mannerthat LTE and NR can evolve in complementary ways As itmight be envisioned a 5G system entails devices connectedto a 5G access network which in turn is connected to a5G core network The 5G access network may include 3GPP(3rd Generation Partnership Project) radio base stationsandor a non-3GPP access network The 5G core networkoffers major improvements compared with a 4G system inthe area of network slicing and service-based architectures(SBAs) in particular the core is designed to support cloudimplementation and the IoT 5G systems subsume important4G system concepts such as the energy saving capabilitiesof narrowband IoT (NB-IoT) radios secure low latencysmall data transmission for low-power devices -- low latencyis a requirement for making autonomous vehicles safe --and devices using energy-preserving dormant states whenpossible Network slicing allows service providers to deliverldquoNetwork as a Service (NaaS)rdquo to largeinstitutional usersaffording them the flexibility to manage their own servicesand devices on the 5G providerrsquos network
Applications driving wireless traffic include but arenot limited to on-demand mobile information and high-resolution entertainment augmented reality virtual realityand immersive services e-health and ubiquitous IoT roll-outsWhile 5G technology could still take several distinct ser-vice directions it appears at this juncture that the view favor-ing a super-fast mobile network where densely-clusteredsmall cells provide contiguous urban coverage to mobile aswell as stationary users is the approach envisioned by thestandards development bodies and by the implementers Itshould be noted that in the US upwards of 55 percent ofresidential users now utilize cellular-services-only at home inplace of a landline and about 30 percent of residential usersutilize both with the trend favoring an eventual transitionto the former Therefore the evolving 5G systems will haveto properly support this growing segment of the market Agoal of 5G networks is to be five times as fast as comparedto the highest current speed of existing 4G networks withdownload speeds as high as 5 Gbps ndash 4G offering only up to amaximum of 1 Gbps Deployment of 5G networks started in2018 in some advanced countries although further develop-ments on fundamentals will continue naturally the current
4 Wireless Communications and Mobile Computing
Table1Ke
yUrban
Challenges
andIoT-supp
ortedSolutio
ns
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Infrastructureandrealestate
managem
ent
Requ
irementmon
itorstatusa
ndoccupancyo
fspacesbu
ildingsroads
bridgestunn
elsrailroadcrossin
gsand
streetsignals
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)toprovider
eal-tim
eand
histo
ricaltre
ndingdataallowingcity
agencies
toprovidee
nhancedvisib
ility
into
thep
erform
ance
ofresources
facilitatingenvironm
entaland
safety
sensingsm
artp
arking
andsm
artp
arking
meterssm
artelectric
metersandsm
art
build
ingfunctio
nality
YY
High
Low
Low
Medium
Livability
Requ
irementQualityof
Lifeexp
editiou
saccessto
servicesefficienttranspo
rtation
lowdelayssafety
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)tofacilitates
martm
ulti-mod
altransportatio
ninform
ation-ric
henvironm
ents
locatio
n-basedservices
real-timec
onnectivity
tohealth-m
onito
ringresources(eg
air
quality
)
YY
High
Medium
Medium
Medium
Logistics
Requ
irementsupp
lyingcitydw
ellers
with
fresh
food
sup
pliesgood
sand
otherm
aterials
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)toenablethes
tream
liningof
warehou
singtransportatio
nand
distr
ibutionof
good
sTraffi
cmanagem
entisa
faceto
fsuchlogistical
supp
ort
YY
High
MediumM
edium
High
Physicalsecurity
Requ
irementsecurityinstr
eets
parks
statio
nstun
nels
bridgestrainsbuses
ferries
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
esandgu
nsho
tdetectio
nsyste
ms)to
supp
ortIP-basedsurveillancev
ideo
license
plater
eading
gun
-sho
tdetectio
nbio-hazard
andradiological
contam
inationmon
itorin
gface
recogn
ition
and
crow
dmon
itorin
gand
control
Perhaps
YHigh
High
Low
High
Wireless Communications and Mobile Computing 5
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Powe
rand
otherc
ity-sup
portingutilitie
s
Requ
irementreliablefl
owof
electric
energygasand
wateroptim
ized
waste-m
anagem
entand
sewe
rsafe
storage
ofgasolin
e
SmartG
ridsolutio
nsandsensor-rich
utilityinfrastructure
NY
High
Low
Medium
High
Traffi
ctransportatio
nandmob
ility
Requ
irementop
timized
traffi
cflow
low
congestio
nlowlatencya
ndhigh
expediencylow
noise
minim
alwasteof
fuelandCO2em
issionssafety
Netwo
rked
sensorstosupp
orttrafficfl
ow
driverlessvehiclesinclu
ding
driverless
bustransit
andmulti-mod
altransportatio
nsyste
msFo
rdriv
erless
vehicles
sensorsw
illallow
high
-resolutionmapping
telem
etry
data
traffi
cand
hazard
avoidancem
echanism
s
NY
High
Medium-to
-High
Low
High
Electricandotheru
tility
manho
lemon
itorin
g
Requ
irementElectricpo
werm
anho
les
requ
iremon
itorin
gto
avoidandor
preventd
angerous
situatio
ns
Cost-e
ffectivea
ndreliables
ensorsare
neededTechn
olog
ybeing
investigatedby
Con
Ediso
nin
New
York
city
NY
High
Low
Medium
High
Pollu
tionmon
itorin
g
Requ
irementmon
itore
missionof
dioxinsvapo
rized
mercury
nano
particlesradiationfro
mfactories
incineratorsurban
crem
atoriaespecially
iftheses
ources
arec
lose
totraintracks
orotherw
ind-turbulence
elem
ents(eg
canyon
s)
Netwo
rked
sensorsthrou
ghou
ttow
n(or
with
in10
kmof
apoint
source)to
mon
itortoxichealth
-impactingem
ission
from
pointsou
rces
inclu
ding
factories
generatio
nplants(if
any)
andcrem
atoria
(ifany)
[35ndash46
]
NY
High
MediumM
edium
High
6 Wireless Communications and Mobile Computing
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Environm
entalM
onito
ring
Requ
irements
mon
itoro
utdo
ortemperaturehum
idity
andother
environm
entalgases
Sensorstothatcanbe
placed
ineasy-to
-deploylocatio
nsegatop
existingSm
artC
itylig
htpo
lesto
continuo
uslymon
itortem
perature
humidity
andothere
nviro
nmentalgases
NY
High
Low
MediumM
edium
Floo
dAb
atem
ent
Requ
irementFloo
dandsto
rmdrainage
control
Distrib
uted
ruggedized
sensorsto
mon
itorF
lood
andsto
rmdrainage
toprovidee
arlywarning
andfaultd
etectio
nN
YHigh
Low
Medium
High
SmartC
ityLigh
ting
Requ
irementCon
versionto
LED
lightingandensuingcontrolviaIoTfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
Citie
sspend
largea
mou
ntso
fmon
eyyearlyforstre
etlig
hting(usually1000
streetlightsp
er10000
inhabitantsand
$125
pery
earp
erlig
htfor4
662ho
urso
fusagey
early
andsyste
mam
ortization)
LEDlig
htingrequ
ires13rd
thea
mou
ntof
powe
rfor
thes
amea
mou
ntof
luminance
Paybackforc
onversionisno
warou
nd5-6
yearsSensorsa
reneeded
for
IoT-directed
light
managem
entfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
NY
High
Medium
Medium
Medium
Wireless Communications and Mobile Computing 7
Table2Ke
yWire
lessTechno
logies
applicableto
IoT
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
5GYesperhapsw
ithDistrib
uted
Antenna
Syste
ms(DASs)
Yesabou
t10-15
miles
(i)Evolving
not
yetw
idely
deployed
(ii)S
everalband
slowlatencyhigh
sensor
density
(iii)Cellularn
etwo
rkarchitecture
(iv)L
icensedspectrum
001M
bpsinsomeimplem
entatio
nsbattery
sim10years
(v)B
roadband
features
availablefor
surveillancemultim
edia
(vi)Cost-e
ffective
(vii)
Expected
tobe
availablew
orldwide
(viii)B
uildingpenetrationmay
need
Distrib
uted
Antenna
Syste
ms
(DASs)
NB-IoT
(Narrowband
IoT)
Yes
Yesup
toabou
t20m
iles
(i)Severalbandslicensedspectrum
(ii)L
TE-based
(iii)01-0
2Mbp
sdatar
atesbatterysim10
+years
(iv)L
owcost
lowmod
emcomplexitylow
powe
renergy
saving
mechanism
s(high
batte
rylife)
(v)D
oesn
otrequ
ireag
atew
aysensord
ataissentd
irectlyto
the
destinatio
nserver
(other
IoTsyste
mstypicallyhave
gatewaysthat
aggregates
ensord
atawhich
then
commun
icatew
iththed
estin
ation
server)
(vi)Re
ason
ablebu
ildingpenetration(im
proved
indo
orcoverage)
(vii)
Largen
umbero
flow
throug
hput
devices(up
to15000
0devices
perc
ell)
8 Wireless Communications and Mobile Computing
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
LTE-M
(Lon
g-Term
EvolutionMachine
Type
Com
mun
ications)
Rel13(C
atM1C
atM)
Yes
Yesabou
t10-20
miles
(i)Cellularn
etwo
rkarchitectureLT
Ecompatib
leeasyto
deployn
ewcellu
lara
ntennasn
otrequ
ired
(ii)U
ses4
G-LTE
band
sbelow
1GHzlicensedspectrum
(iii)Con
sidered
thes
econ
dgeneratio
nof
LTEchipsa
imed
atIoT
applications
(iv)C
apsm
axim
umsyste
mband
width
at14
MHz(
asop
posedto
Cat-0rsquos20
MHz)thu
sisc
ost-e
ffectivefor
LowPo
werW
ideA
rea
Netwo
rk(LPW
AN)app
lications
such
assm
artm
eteringwhereon
lysm
allamou
ntof
datatransfe
risrequired
(v)1
Mbp
suploaddo
wnload
batte
rysim10
years
(vi)Re
lativ
elylowcomplexity
andlowpo
werm
odem
(vii)
Can
beused
fortrackingmovingob
jects(Lo
catio
nservices
provided
throug
hcelltowe
rmechanism
s)
LoRa
Yes
Yes(6-15
milesw
ithLO
S)
(i)Ba
ndbelow1G
Hz
(ii)IoT
-focusedfro
mtheg
et-go
(iii)Prop
rietary
(iv)L
owpo
wer
Sigfox
Somew
hatlim
ited
Yes(30
milesinrural
environm
ents
1-6miles
incityenvironm
ents)
(i)Ba
ndbelow1G
Hz
(ii)N
arrowband
(iii)Lo
wpo
wer
(iv)S
tartop
olog
y
Wireless Communications and Mobile Computing 9
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
Wi-F
iYes300feet
Tosomed
egreerequ
ires
inter-spot
conn
ectiv
itybackbo
ne(w
iredor
wire
less)(eg
80211ah
dista
ncer
ange
upto
abou
t12
mile)
(i)Severalbands
(ii)In2018
theF
CCallowe
dthee
xpansio
nof
the6
GHzb
andto
next-generationWi-F
idevices
with
12GHzo
fadd
ition
alspectrum
spanning
5925to
7125
GHz(
currentW
i-Fin
etwo
rkso
perateat24
GHza
nd5GHzw
ithafew
vend
orso
fferin
g60
GHzldquo
WiGigrdquothis
having
arange
of30
feetndashIEEE
80211a
dandIEEE
80211a
y)(iii)Highadop
tion
most(bu
tnot
all)indo
orIoTutilize
Wi-F
igood
functio
nality
(iv)F
reeldquo
airtim
erdquo(v)S
ubjectto
interfe
rencemalicious
orno
n-malicious
interfe
rence
(egtoo
manyho
tspo
ts)couldim
pairthes
ensorfrom
send
ingdata
either
onafi
ne-grain
orcoarse-grain
basis
Bluetooth
Yes30
feet
No(orfor
Person
alArea
only)
(i)Lo
wband
width
(2Mbp
s)(ii)U
sedin
medicaldevicesa
ndindu
strialsensorsLo
wpo
wergood
forw
earables
(iii)Usablefor
Realtim
elocationsyste
msw
ithmedium
accuracy
Zigbee
Yes(30-300
feet)
No(orfor
Person
alArea
only)
(i)Lo
wdatarate
(ii)Ind
ustrialand
someh
omea
pplications
(egho
mee
nergy
mon
itorin
gwire
lesslig
htsw
itches)
(iii)Lo
wtransm
itpo
werLo
wbatte
ryconsum
ption
NoteAfewotherlegacyIoTwire
lesstechno
logies
exist
(egCat0Cat1EC
-GSM
Weightless)b
utaren
otinclu
dedin
thistable
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
4 Wireless Communications and Mobile Computing
Table1Ke
yUrban
Challenges
andIoT-supp
ortedSolutio
ns
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Infrastructureandrealestate
managem
ent
Requ
irementmon
itorstatusa
ndoccupancyo
fspacesbu
ildingsroads
bridgestunn
elsrailroadcrossin
gsand
streetsignals
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)toprovider
eal-tim
eand
histo
ricaltre
ndingdataallowingcity
agencies
toprovidee
nhancedvisib
ility
into
thep
erform
ance
ofresources
facilitatingenvironm
entaland
safety
sensingsm
artp
arking
andsm
artp
arking
meterssm
artelectric
metersandsm
art
build
ingfunctio
nality
YY
High
Low
Low
Medium
Livability
Requ
irementQualityof
Lifeexp
editiou
saccessto
servicesefficienttranspo
rtation
lowdelayssafety
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)tofacilitates
martm
ulti-mod
altransportatio
ninform
ation-ric
henvironm
ents
locatio
n-basedservices
real-timec
onnectivity
tohealth-m
onito
ringresources(eg
air
quality
)
YY
High
Medium
Medium
Medium
Logistics
Requ
irementsupp
lyingcitydw
ellers
with
fresh
food
sup
pliesgood
sand
otherm
aterials
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
es)toenablethes
tream
liningof
warehou
singtransportatio
nand
distr
ibutionof
good
sTraffi
cmanagem
entisa
faceto
fsuchlogistical
supp
ort
YY
High
MediumM
edium
High
Physicalsecurity
Requ
irementsecurityinstr
eets
parks
statio
nstun
nels
bridgestrainsbuses
ferries
Netwo
rked
sensors(po
ssiblyinclu
ding
dron
esandgu
nsho
tdetectio
nsyste
ms)to
supp
ortIP-basedsurveillancev
ideo
license
plater
eading
gun
-sho
tdetectio
nbio-hazard
andradiological
contam
inationmon
itorin
gface
recogn
ition
and
crow
dmon
itorin
gand
control
Perhaps
YHigh
High
Low
High
Wireless Communications and Mobile Computing 5
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Powe
rand
otherc
ity-sup
portingutilitie
s
Requ
irementreliablefl
owof
electric
energygasand
wateroptim
ized
waste-m
anagem
entand
sewe
rsafe
storage
ofgasolin
e
SmartG
ridsolutio
nsandsensor-rich
utilityinfrastructure
NY
High
Low
Medium
High
Traffi
ctransportatio
nandmob
ility
Requ
irementop
timized
traffi
cflow
low
congestio
nlowlatencya
ndhigh
expediencylow
noise
minim
alwasteof
fuelandCO2em
issionssafety
Netwo
rked
sensorstosupp
orttrafficfl
ow
driverlessvehiclesinclu
ding
driverless
bustransit
andmulti-mod
altransportatio
nsyste
msFo
rdriv
erless
vehicles
sensorsw
illallow
high
-resolutionmapping
telem
etry
data
traffi
cand
hazard
avoidancem
echanism
s
NY
High
Medium-to
-High
Low
High
Electricandotheru
tility
manho
lemon
itorin
g
Requ
irementElectricpo
werm
anho
les
requ
iremon
itorin
gto
avoidandor
preventd
angerous
situatio
ns
Cost-e
ffectivea
ndreliables
ensorsare
neededTechn
olog
ybeing
investigatedby
Con
Ediso
nin
New
York
city
NY
High
Low
Medium
High
Pollu
tionmon
itorin
g
Requ
irementmon
itore
missionof
dioxinsvapo
rized
mercury
nano
particlesradiationfro
mfactories
incineratorsurban
crem
atoriaespecially
iftheses
ources
arec
lose
totraintracks
orotherw
ind-turbulence
elem
ents(eg
canyon
s)
Netwo
rked
sensorsthrou
ghou
ttow
n(or
with
in10
kmof
apoint
source)to
mon
itortoxichealth
-impactingem
ission
from
pointsou
rces
inclu
ding
factories
generatio
nplants(if
any)
andcrem
atoria
(ifany)
[35ndash46
]
NY
High
MediumM
edium
High
6 Wireless Communications and Mobile Computing
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Environm
entalM
onito
ring
Requ
irements
mon
itoro
utdo
ortemperaturehum
idity
andother
environm
entalgases
Sensorstothatcanbe
placed
ineasy-to
-deploylocatio
nsegatop
existingSm
artC
itylig
htpo
lesto
continuo
uslymon
itortem
perature
humidity
andothere
nviro
nmentalgases
NY
High
Low
MediumM
edium
Floo
dAb
atem
ent
Requ
irementFloo
dandsto
rmdrainage
control
Distrib
uted
ruggedized
sensorsto
mon
itorF
lood
andsto
rmdrainage
toprovidee
arlywarning
andfaultd
etectio
nN
YHigh
Low
Medium
High
SmartC
ityLigh
ting
Requ
irementCon
versionto
LED
lightingandensuingcontrolviaIoTfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
Citie
sspend
largea
mou
ntso
fmon
eyyearlyforstre
etlig
hting(usually1000
streetlightsp
er10000
inhabitantsand
$125
pery
earp
erlig
htfor4
662ho
urso
fusagey
early
andsyste
mam
ortization)
LEDlig
htingrequ
ires13rd
thea
mou
ntof
powe
rfor
thes
amea
mou
ntof
luminance
Paybackforc
onversionisno
warou
nd5-6
yearsSensorsa
reneeded
for
IoT-directed
light
managem
entfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
NY
High
Medium
Medium
Medium
Wireless Communications and Mobile Computing 7
Table2Ke
yWire
lessTechno
logies
applicableto
IoT
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
5GYesperhapsw
ithDistrib
uted
Antenna
Syste
ms(DASs)
Yesabou
t10-15
miles
(i)Evolving
not
yetw
idely
deployed
(ii)S
everalband
slowlatencyhigh
sensor
density
(iii)Cellularn
etwo
rkarchitecture
(iv)L
icensedspectrum
001M
bpsinsomeimplem
entatio
nsbattery
sim10years
(v)B
roadband
features
availablefor
surveillancemultim
edia
(vi)Cost-e
ffective
(vii)
Expected
tobe
availablew
orldwide
(viii)B
uildingpenetrationmay
need
Distrib
uted
Antenna
Syste
ms
(DASs)
NB-IoT
(Narrowband
IoT)
Yes
Yesup
toabou
t20m
iles
(i)Severalbandslicensedspectrum
(ii)L
TE-based
(iii)01-0
2Mbp
sdatar
atesbatterysim10
+years
(iv)L
owcost
lowmod
emcomplexitylow
powe
renergy
saving
mechanism
s(high
batte
rylife)
(v)D
oesn
otrequ
ireag
atew
aysensord
ataissentd
irectlyto
the
destinatio
nserver
(other
IoTsyste
mstypicallyhave
gatewaysthat
aggregates
ensord
atawhich
then
commun
icatew
iththed
estin
ation
server)
(vi)Re
ason
ablebu
ildingpenetration(im
proved
indo
orcoverage)
(vii)
Largen
umbero
flow
throug
hput
devices(up
to15000
0devices
perc
ell)
8 Wireless Communications and Mobile Computing
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
LTE-M
(Lon
g-Term
EvolutionMachine
Type
Com
mun
ications)
Rel13(C
atM1C
atM)
Yes
Yesabou
t10-20
miles
(i)Cellularn
etwo
rkarchitectureLT
Ecompatib
leeasyto
deployn
ewcellu
lara
ntennasn
otrequ
ired
(ii)U
ses4
G-LTE
band
sbelow
1GHzlicensedspectrum
(iii)Con
sidered
thes
econ
dgeneratio
nof
LTEchipsa
imed
atIoT
applications
(iv)C
apsm
axim
umsyste
mband
width
at14
MHz(
asop
posedto
Cat-0rsquos20
MHz)thu
sisc
ost-e
ffectivefor
LowPo
werW
ideA
rea
Netwo
rk(LPW
AN)app
lications
such
assm
artm
eteringwhereon
lysm
allamou
ntof
datatransfe
risrequired
(v)1
Mbp
suploaddo
wnload
batte
rysim10
years
(vi)Re
lativ
elylowcomplexity
andlowpo
werm
odem
(vii)
Can
beused
fortrackingmovingob
jects(Lo
catio
nservices
provided
throug
hcelltowe
rmechanism
s)
LoRa
Yes
Yes(6-15
milesw
ithLO
S)
(i)Ba
ndbelow1G
Hz
(ii)IoT
-focusedfro
mtheg
et-go
(iii)Prop
rietary
(iv)L
owpo
wer
Sigfox
Somew
hatlim
ited
Yes(30
milesinrural
environm
ents
1-6miles
incityenvironm
ents)
(i)Ba
ndbelow1G
Hz
(ii)N
arrowband
(iii)Lo
wpo
wer
(iv)S
tartop
olog
y
Wireless Communications and Mobile Computing 9
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
Wi-F
iYes300feet
Tosomed
egreerequ
ires
inter-spot
conn
ectiv
itybackbo
ne(w
iredor
wire
less)(eg
80211ah
dista
ncer
ange
upto
abou
t12
mile)
(i)Severalbands
(ii)In2018
theF
CCallowe
dthee
xpansio
nof
the6
GHzb
andto
next-generationWi-F
idevices
with
12GHzo
fadd
ition
alspectrum
spanning
5925to
7125
GHz(
currentW
i-Fin
etwo
rkso
perateat24
GHza
nd5GHzw
ithafew
vend
orso
fferin
g60
GHzldquo
WiGigrdquothis
having
arange
of30
feetndashIEEE
80211a
dandIEEE
80211a
y)(iii)Highadop
tion
most(bu
tnot
all)indo
orIoTutilize
Wi-F
igood
functio
nality
(iv)F
reeldquo
airtim
erdquo(v)S
ubjectto
interfe
rencemalicious
orno
n-malicious
interfe
rence
(egtoo
manyho
tspo
ts)couldim
pairthes
ensorfrom
send
ingdata
either
onafi
ne-grain
orcoarse-grain
basis
Bluetooth
Yes30
feet
No(orfor
Person
alArea
only)
(i)Lo
wband
width
(2Mbp
s)(ii)U
sedin
medicaldevicesa
ndindu
strialsensorsLo
wpo
wergood
forw
earables
(iii)Usablefor
Realtim
elocationsyste
msw
ithmedium
accuracy
Zigbee
Yes(30-300
feet)
No(orfor
Person
alArea
only)
(i)Lo
wdatarate
(ii)Ind
ustrialand
someh
omea
pplications
(egho
mee
nergy
mon
itorin
gwire
lesslig
htsw
itches)
(iii)Lo
wtransm
itpo
werLo
wbatte
ryconsum
ption
NoteAfewotherlegacyIoTwire
lesstechno
logies
exist
(egCat0Cat1EC
-GSM
Weightless)b
utaren
otinclu
dedin
thistable
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 5
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Powe
rand
otherc
ity-sup
portingutilitie
s
Requ
irementreliablefl
owof
electric
energygasand
wateroptim
ized
waste-m
anagem
entand
sewe
rsafe
storage
ofgasolin
e
SmartG
ridsolutio
nsandsensor-rich
utilityinfrastructure
NY
High
Low
Medium
High
Traffi
ctransportatio
nandmob
ility
Requ
irementop
timized
traffi
cflow
low
congestio
nlowlatencya
ndhigh
expediencylow
noise
minim
alwasteof
fuelandCO2em
issionssafety
Netwo
rked
sensorstosupp
orttrafficfl
ow
driverlessvehiclesinclu
ding
driverless
bustransit
andmulti-mod
altransportatio
nsyste
msFo
rdriv
erless
vehicles
sensorsw
illallow
high
-resolutionmapping
telem
etry
data
traffi
cand
hazard
avoidancem
echanism
s
NY
High
Medium-to
-High
Low
High
Electricandotheru
tility
manho
lemon
itorin
g
Requ
irementElectricpo
werm
anho
les
requ
iremon
itorin
gto
avoidandor
preventd
angerous
situatio
ns
Cost-e
ffectivea
ndreliables
ensorsare
neededTechn
olog
ybeing
investigatedby
Con
Ediso
nin
New
York
city
NY
High
Low
Medium
High
Pollu
tionmon
itorin
g
Requ
irementmon
itore
missionof
dioxinsvapo
rized
mercury
nano
particlesradiationfro
mfactories
incineratorsurban
crem
atoriaespecially
iftheses
ources
arec
lose
totraintracks
orotherw
ind-turbulence
elem
ents(eg
canyon
s)
Netwo
rked
sensorsthrou
ghou
ttow
n(or
with
in10
kmof
apoint
source)to
mon
itortoxichealth
-impactingem
ission
from
pointsou
rces
inclu
ding
factories
generatio
nplants(if
any)
andcrem
atoria
(ifany)
[35ndash46
]
NY
High
MediumM
edium
High
6 Wireless Communications and Mobile Computing
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Environm
entalM
onito
ring
Requ
irements
mon
itoro
utdo
ortemperaturehum
idity
andother
environm
entalgases
Sensorstothatcanbe
placed
ineasy-to
-deploylocatio
nsegatop
existingSm
artC
itylig
htpo
lesto
continuo
uslymon
itortem
perature
humidity
andothere
nviro
nmentalgases
NY
High
Low
MediumM
edium
Floo
dAb
atem
ent
Requ
irementFloo
dandsto
rmdrainage
control
Distrib
uted
ruggedized
sensorsto
mon
itorF
lood
andsto
rmdrainage
toprovidee
arlywarning
andfaultd
etectio
nN
YHigh
Low
Medium
High
SmartC
ityLigh
ting
Requ
irementCon
versionto
LED
lightingandensuingcontrolviaIoTfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
Citie
sspend
largea
mou
ntso
fmon
eyyearlyforstre
etlig
hting(usually1000
streetlightsp
er10000
inhabitantsand
$125
pery
earp
erlig
htfor4
662ho
urso
fusagey
early
andsyste
mam
ortization)
LEDlig
htingrequ
ires13rd
thea
mou
ntof
powe
rfor
thes
amea
mou
ntof
luminance
Paybackforc
onversionisno
warou
nd5-6
yearsSensorsa
reneeded
for
IoT-directed
light
managem
entfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
NY
High
Medium
Medium
Medium
Wireless Communications and Mobile Computing 7
Table2Ke
yWire
lessTechno
logies
applicableto
IoT
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
5GYesperhapsw
ithDistrib
uted
Antenna
Syste
ms(DASs)
Yesabou
t10-15
miles
(i)Evolving
not
yetw
idely
deployed
(ii)S
everalband
slowlatencyhigh
sensor
density
(iii)Cellularn
etwo
rkarchitecture
(iv)L
icensedspectrum
001M
bpsinsomeimplem
entatio
nsbattery
sim10years
(v)B
roadband
features
availablefor
surveillancemultim
edia
(vi)Cost-e
ffective
(vii)
Expected
tobe
availablew
orldwide
(viii)B
uildingpenetrationmay
need
Distrib
uted
Antenna
Syste
ms
(DASs)
NB-IoT
(Narrowband
IoT)
Yes
Yesup
toabou
t20m
iles
(i)Severalbandslicensedspectrum
(ii)L
TE-based
(iii)01-0
2Mbp
sdatar
atesbatterysim10
+years
(iv)L
owcost
lowmod
emcomplexitylow
powe
renergy
saving
mechanism
s(high
batte
rylife)
(v)D
oesn
otrequ
ireag
atew
aysensord
ataissentd
irectlyto
the
destinatio
nserver
(other
IoTsyste
mstypicallyhave
gatewaysthat
aggregates
ensord
atawhich
then
commun
icatew
iththed
estin
ation
server)
(vi)Re
ason
ablebu
ildingpenetration(im
proved
indo
orcoverage)
(vii)
Largen
umbero
flow
throug
hput
devices(up
to15000
0devices
perc
ell)
8 Wireless Communications and Mobile Computing
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
LTE-M
(Lon
g-Term
EvolutionMachine
Type
Com
mun
ications)
Rel13(C
atM1C
atM)
Yes
Yesabou
t10-20
miles
(i)Cellularn
etwo
rkarchitectureLT
Ecompatib
leeasyto
deployn
ewcellu
lara
ntennasn
otrequ
ired
(ii)U
ses4
G-LTE
band
sbelow
1GHzlicensedspectrum
(iii)Con
sidered
thes
econ
dgeneratio
nof
LTEchipsa
imed
atIoT
applications
(iv)C
apsm
axim
umsyste
mband
width
at14
MHz(
asop
posedto
Cat-0rsquos20
MHz)thu
sisc
ost-e
ffectivefor
LowPo
werW
ideA
rea
Netwo
rk(LPW
AN)app
lications
such
assm
artm
eteringwhereon
lysm
allamou
ntof
datatransfe
risrequired
(v)1
Mbp
suploaddo
wnload
batte
rysim10
years
(vi)Re
lativ
elylowcomplexity
andlowpo
werm
odem
(vii)
Can
beused
fortrackingmovingob
jects(Lo
catio
nservices
provided
throug
hcelltowe
rmechanism
s)
LoRa
Yes
Yes(6-15
milesw
ithLO
S)
(i)Ba
ndbelow1G
Hz
(ii)IoT
-focusedfro
mtheg
et-go
(iii)Prop
rietary
(iv)L
owpo
wer
Sigfox
Somew
hatlim
ited
Yes(30
milesinrural
environm
ents
1-6miles
incityenvironm
ents)
(i)Ba
ndbelow1G
Hz
(ii)N
arrowband
(iii)Lo
wpo
wer
(iv)S
tartop
olog
y
Wireless Communications and Mobile Computing 9
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
Wi-F
iYes300feet
Tosomed
egreerequ
ires
inter-spot
conn
ectiv
itybackbo
ne(w
iredor
wire
less)(eg
80211ah
dista
ncer
ange
upto
abou
t12
mile)
(i)Severalbands
(ii)In2018
theF
CCallowe
dthee
xpansio
nof
the6
GHzb
andto
next-generationWi-F
idevices
with
12GHzo
fadd
ition
alspectrum
spanning
5925to
7125
GHz(
currentW
i-Fin
etwo
rkso
perateat24
GHza
nd5GHzw
ithafew
vend
orso
fferin
g60
GHzldquo
WiGigrdquothis
having
arange
of30
feetndashIEEE
80211a
dandIEEE
80211a
y)(iii)Highadop
tion
most(bu
tnot
all)indo
orIoTutilize
Wi-F
igood
functio
nality
(iv)F
reeldquo
airtim
erdquo(v)S
ubjectto
interfe
rencemalicious
orno
n-malicious
interfe
rence
(egtoo
manyho
tspo
ts)couldim
pairthes
ensorfrom
send
ingdata
either
onafi
ne-grain
orcoarse-grain
basis
Bluetooth
Yes30
feet
No(orfor
Person
alArea
only)
(i)Lo
wband
width
(2Mbp
s)(ii)U
sedin
medicaldevicesa
ndindu
strialsensorsLo
wpo
wergood
forw
earables
(iii)Usablefor
Realtim
elocationsyste
msw
ithmedium
accuracy
Zigbee
Yes(30-300
feet)
No(orfor
Person
alArea
only)
(i)Lo
wdatarate
(ii)Ind
ustrialand
someh
omea
pplications
(egho
mee
nergy
mon
itorin
gwire
lesslig
htsw
itches)
(iii)Lo
wtransm
itpo
werLo
wbatte
ryconsum
ption
NoteAfewotherlegacyIoTwire
lesstechno
logies
exist
(egCat0Cat1EC
-GSM
Weightless)b
utaren
otinclu
dedin
thistable
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
6 Wireless Communications and Mobile Computing
Table1Con
tinued
SmartC
ityIss
ueandRe
quire
ments
IoTsupp
orts
olutions
Indo
ors
wire
less
needed
Outdo
ors
wire
less
needed
5Gapplicability
Band
width
latency
reliability
Environm
entalM
onito
ring
Requ
irements
mon
itoro
utdo
ortemperaturehum
idity
andother
environm
entalgases
Sensorstothatcanbe
placed
ineasy-to
-deploylocatio
nsegatop
existingSm
artC
itylig
htpo
lesto
continuo
uslymon
itortem
perature
humidity
andothere
nviro
nmentalgases
NY
High
Low
MediumM
edium
Floo
dAb
atem
ent
Requ
irementFloo
dandsto
rmdrainage
control
Distrib
uted
ruggedized
sensorsto
mon
itorF
lood
andsto
rmdrainage
toprovidee
arlywarning
andfaultd
etectio
nN
YHigh
Low
Medium
High
SmartC
ityLigh
ting
Requ
irementCon
versionto
LED
lightingandensuingcontrolviaIoTfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
Citie
sspend
largea
mou
ntso
fmon
eyyearlyforstre
etlig
hting(usually1000
streetlightsp
er10000
inhabitantsand
$125
pery
earp
erlig
htfor4
662ho
urso
fusagey
early
andsyste
mam
ortization)
LEDlig
htingrequ
ires13rd
thea
mou
ntof
powe
rfor
thes
amea
mou
ntof
luminance
Paybackforc
onversionisno
warou
nd5-6
yearsSensorsa
reneeded
for
IoT-directed
light
managem
entfor
weatherc
onditio
nsphaseso
fthe
moo
nseason
straffi
coccup
ancyand
soon
NY
High
Medium
Medium
Medium
Wireless Communications and Mobile Computing 7
Table2Ke
yWire
lessTechno
logies
applicableto
IoT
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
5GYesperhapsw
ithDistrib
uted
Antenna
Syste
ms(DASs)
Yesabou
t10-15
miles
(i)Evolving
not
yetw
idely
deployed
(ii)S
everalband
slowlatencyhigh
sensor
density
(iii)Cellularn
etwo
rkarchitecture
(iv)L
icensedspectrum
001M
bpsinsomeimplem
entatio
nsbattery
sim10years
(v)B
roadband
features
availablefor
surveillancemultim
edia
(vi)Cost-e
ffective
(vii)
Expected
tobe
availablew
orldwide
(viii)B
uildingpenetrationmay
need
Distrib
uted
Antenna
Syste
ms
(DASs)
NB-IoT
(Narrowband
IoT)
Yes
Yesup
toabou
t20m
iles
(i)Severalbandslicensedspectrum
(ii)L
TE-based
(iii)01-0
2Mbp
sdatar
atesbatterysim10
+years
(iv)L
owcost
lowmod
emcomplexitylow
powe
renergy
saving
mechanism
s(high
batte
rylife)
(v)D
oesn
otrequ
ireag
atew
aysensord
ataissentd
irectlyto
the
destinatio
nserver
(other
IoTsyste
mstypicallyhave
gatewaysthat
aggregates
ensord
atawhich
then
commun
icatew
iththed
estin
ation
server)
(vi)Re
ason
ablebu
ildingpenetration(im
proved
indo
orcoverage)
(vii)
Largen
umbero
flow
throug
hput
devices(up
to15000
0devices
perc
ell)
8 Wireless Communications and Mobile Computing
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
LTE-M
(Lon
g-Term
EvolutionMachine
Type
Com
mun
ications)
Rel13(C
atM1C
atM)
Yes
Yesabou
t10-20
miles
(i)Cellularn
etwo
rkarchitectureLT
Ecompatib
leeasyto
deployn
ewcellu
lara
ntennasn
otrequ
ired
(ii)U
ses4
G-LTE
band
sbelow
1GHzlicensedspectrum
(iii)Con
sidered
thes
econ
dgeneratio
nof
LTEchipsa
imed
atIoT
applications
(iv)C
apsm
axim
umsyste
mband
width
at14
MHz(
asop
posedto
Cat-0rsquos20
MHz)thu
sisc
ost-e
ffectivefor
LowPo
werW
ideA
rea
Netwo
rk(LPW
AN)app
lications
such
assm
artm
eteringwhereon
lysm
allamou
ntof
datatransfe
risrequired
(v)1
Mbp
suploaddo
wnload
batte
rysim10
years
(vi)Re
lativ
elylowcomplexity
andlowpo
werm
odem
(vii)
Can
beused
fortrackingmovingob
jects(Lo
catio
nservices
provided
throug
hcelltowe
rmechanism
s)
LoRa
Yes
Yes(6-15
milesw
ithLO
S)
(i)Ba
ndbelow1G
Hz
(ii)IoT
-focusedfro
mtheg
et-go
(iii)Prop
rietary
(iv)L
owpo
wer
Sigfox
Somew
hatlim
ited
Yes(30
milesinrural
environm
ents
1-6miles
incityenvironm
ents)
(i)Ba
ndbelow1G
Hz
(ii)N
arrowband
(iii)Lo
wpo
wer
(iv)S
tartop
olog
y
Wireless Communications and Mobile Computing 9
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
Wi-F
iYes300feet
Tosomed
egreerequ
ires
inter-spot
conn
ectiv
itybackbo
ne(w
iredor
wire
less)(eg
80211ah
dista
ncer
ange
upto
abou
t12
mile)
(i)Severalbands
(ii)In2018
theF
CCallowe
dthee
xpansio
nof
the6
GHzb
andto
next-generationWi-F
idevices
with
12GHzo
fadd
ition
alspectrum
spanning
5925to
7125
GHz(
currentW
i-Fin
etwo
rkso
perateat24
GHza
nd5GHzw
ithafew
vend
orso
fferin
g60
GHzldquo
WiGigrdquothis
having
arange
of30
feetndashIEEE
80211a
dandIEEE
80211a
y)(iii)Highadop
tion
most(bu
tnot
all)indo
orIoTutilize
Wi-F
igood
functio
nality
(iv)F
reeldquo
airtim
erdquo(v)S
ubjectto
interfe
rencemalicious
orno
n-malicious
interfe
rence
(egtoo
manyho
tspo
ts)couldim
pairthes
ensorfrom
send
ingdata
either
onafi
ne-grain
orcoarse-grain
basis
Bluetooth
Yes30
feet
No(orfor
Person
alArea
only)
(i)Lo
wband
width
(2Mbp
s)(ii)U
sedin
medicaldevicesa
ndindu
strialsensorsLo
wpo
wergood
forw
earables
(iii)Usablefor
Realtim
elocationsyste
msw
ithmedium
accuracy
Zigbee
Yes(30-300
feet)
No(orfor
Person
alArea
only)
(i)Lo
wdatarate
(ii)Ind
ustrialand
someh
omea
pplications
(egho
mee
nergy
mon
itorin
gwire
lesslig
htsw
itches)
(iii)Lo
wtransm
itpo
werLo
wbatte
ryconsum
ption
NoteAfewotherlegacyIoTwire
lesstechno
logies
exist
(egCat0Cat1EC
-GSM
Weightless)b
utaren
otinclu
dedin
thistable
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 7
Table2Ke
yWire
lessTechno
logies
applicableto
IoT
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
5GYesperhapsw
ithDistrib
uted
Antenna
Syste
ms(DASs)
Yesabou
t10-15
miles
(i)Evolving
not
yetw
idely
deployed
(ii)S
everalband
slowlatencyhigh
sensor
density
(iii)Cellularn
etwo
rkarchitecture
(iv)L
icensedspectrum
001M
bpsinsomeimplem
entatio
nsbattery
sim10years
(v)B
roadband
features
availablefor
surveillancemultim
edia
(vi)Cost-e
ffective
(vii)
Expected
tobe
availablew
orldwide
(viii)B
uildingpenetrationmay
need
Distrib
uted
Antenna
Syste
ms
(DASs)
NB-IoT
(Narrowband
IoT)
Yes
Yesup
toabou
t20m
iles
(i)Severalbandslicensedspectrum
(ii)L
TE-based
(iii)01-0
2Mbp
sdatar
atesbatterysim10
+years
(iv)L
owcost
lowmod
emcomplexitylow
powe
renergy
saving
mechanism
s(high
batte
rylife)
(v)D
oesn
otrequ
ireag
atew
aysensord
ataissentd
irectlyto
the
destinatio
nserver
(other
IoTsyste
mstypicallyhave
gatewaysthat
aggregates
ensord
atawhich
then
commun
icatew
iththed
estin
ation
server)
(vi)Re
ason
ablebu
ildingpenetration(im
proved
indo
orcoverage)
(vii)
Largen
umbero
flow
throug
hput
devices(up
to15000
0devices
perc
ell)
8 Wireless Communications and Mobile Computing
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
LTE-M
(Lon
g-Term
EvolutionMachine
Type
Com
mun
ications)
Rel13(C
atM1C
atM)
Yes
Yesabou
t10-20
miles
(i)Cellularn
etwo
rkarchitectureLT
Ecompatib
leeasyto
deployn
ewcellu
lara
ntennasn
otrequ
ired
(ii)U
ses4
G-LTE
band
sbelow
1GHzlicensedspectrum
(iii)Con
sidered
thes
econ
dgeneratio
nof
LTEchipsa
imed
atIoT
applications
(iv)C
apsm
axim
umsyste
mband
width
at14
MHz(
asop
posedto
Cat-0rsquos20
MHz)thu
sisc
ost-e
ffectivefor
LowPo
werW
ideA
rea
Netwo
rk(LPW
AN)app
lications
such
assm
artm
eteringwhereon
lysm
allamou
ntof
datatransfe
risrequired
(v)1
Mbp
suploaddo
wnload
batte
rysim10
years
(vi)Re
lativ
elylowcomplexity
andlowpo
werm
odem
(vii)
Can
beused
fortrackingmovingob
jects(Lo
catio
nservices
provided
throug
hcelltowe
rmechanism
s)
LoRa
Yes
Yes(6-15
milesw
ithLO
S)
(i)Ba
ndbelow1G
Hz
(ii)IoT
-focusedfro
mtheg
et-go
(iii)Prop
rietary
(iv)L
owpo
wer
Sigfox
Somew
hatlim
ited
Yes(30
milesinrural
environm
ents
1-6miles
incityenvironm
ents)
(i)Ba
ndbelow1G
Hz
(ii)N
arrowband
(iii)Lo
wpo
wer
(iv)S
tartop
olog
y
Wireless Communications and Mobile Computing 9
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
Wi-F
iYes300feet
Tosomed
egreerequ
ires
inter-spot
conn
ectiv
itybackbo
ne(w
iredor
wire
less)(eg
80211ah
dista
ncer
ange
upto
abou
t12
mile)
(i)Severalbands
(ii)In2018
theF
CCallowe
dthee
xpansio
nof
the6
GHzb
andto
next-generationWi-F
idevices
with
12GHzo
fadd
ition
alspectrum
spanning
5925to
7125
GHz(
currentW
i-Fin
etwo
rkso
perateat24
GHza
nd5GHzw
ithafew
vend
orso
fferin
g60
GHzldquo
WiGigrdquothis
having
arange
of30
feetndashIEEE
80211a
dandIEEE
80211a
y)(iii)Highadop
tion
most(bu
tnot
all)indo
orIoTutilize
Wi-F
igood
functio
nality
(iv)F
reeldquo
airtim
erdquo(v)S
ubjectto
interfe
rencemalicious
orno
n-malicious
interfe
rence
(egtoo
manyho
tspo
ts)couldim
pairthes
ensorfrom
send
ingdata
either
onafi
ne-grain
orcoarse-grain
basis
Bluetooth
Yes30
feet
No(orfor
Person
alArea
only)
(i)Lo
wband
width
(2Mbp
s)(ii)U
sedin
medicaldevicesa
ndindu
strialsensorsLo
wpo
wergood
forw
earables
(iii)Usablefor
Realtim
elocationsyste
msw
ithmedium
accuracy
Zigbee
Yes(30-300
feet)
No(orfor
Person
alArea
only)
(i)Lo
wdatarate
(ii)Ind
ustrialand
someh
omea
pplications
(egho
mee
nergy
mon
itorin
gwire
lesslig
htsw
itches)
(iii)Lo
wtransm
itpo
werLo
wbatte
ryconsum
ption
NoteAfewotherlegacyIoTwire
lesstechno
logies
exist
(egCat0Cat1EC
-GSM
Weightless)b
utaren
otinclu
dedin
thistable
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
8 Wireless Communications and Mobile Computing
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
LTE-M
(Lon
g-Term
EvolutionMachine
Type
Com
mun
ications)
Rel13(C
atM1C
atM)
Yes
Yesabou
t10-20
miles
(i)Cellularn
etwo
rkarchitectureLT
Ecompatib
leeasyto
deployn
ewcellu
lara
ntennasn
otrequ
ired
(ii)U
ses4
G-LTE
band
sbelow
1GHzlicensedspectrum
(iii)Con
sidered
thes
econ
dgeneratio
nof
LTEchipsa
imed
atIoT
applications
(iv)C
apsm
axim
umsyste
mband
width
at14
MHz(
asop
posedto
Cat-0rsquos20
MHz)thu
sisc
ost-e
ffectivefor
LowPo
werW
ideA
rea
Netwo
rk(LPW
AN)app
lications
such
assm
artm
eteringwhereon
lysm
allamou
ntof
datatransfe
risrequired
(v)1
Mbp
suploaddo
wnload
batte
rysim10
years
(vi)Re
lativ
elylowcomplexity
andlowpo
werm
odem
(vii)
Can
beused
fortrackingmovingob
jects(Lo
catio
nservices
provided
throug
hcelltowe
rmechanism
s)
LoRa
Yes
Yes(6-15
milesw
ithLO
S)
(i)Ba
ndbelow1G
Hz
(ii)IoT
-focusedfro
mtheg
et-go
(iii)Prop
rietary
(iv)L
owpo
wer
Sigfox
Somew
hatlim
ited
Yes(30
milesinrural
environm
ents
1-6miles
incityenvironm
ents)
(i)Ba
ndbelow1G
Hz
(ii)N
arrowband
(iii)Lo
wpo
wer
(iv)S
tartop
olog
y
Wireless Communications and Mobile Computing 9
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
Wi-F
iYes300feet
Tosomed
egreerequ
ires
inter-spot
conn
ectiv
itybackbo
ne(w
iredor
wire
less)(eg
80211ah
dista
ncer
ange
upto
abou
t12
mile)
(i)Severalbands
(ii)In2018
theF
CCallowe
dthee
xpansio
nof
the6
GHzb
andto
next-generationWi-F
idevices
with
12GHzo
fadd
ition
alspectrum
spanning
5925to
7125
GHz(
currentW
i-Fin
etwo
rkso
perateat24
GHza
nd5GHzw
ithafew
vend
orso
fferin
g60
GHzldquo
WiGigrdquothis
having
arange
of30
feetndashIEEE
80211a
dandIEEE
80211a
y)(iii)Highadop
tion
most(bu
tnot
all)indo
orIoTutilize
Wi-F
igood
functio
nality
(iv)F
reeldquo
airtim
erdquo(v)S
ubjectto
interfe
rencemalicious
orno
n-malicious
interfe
rence
(egtoo
manyho
tspo
ts)couldim
pairthes
ensorfrom
send
ingdata
either
onafi
ne-grain
orcoarse-grain
basis
Bluetooth
Yes30
feet
No(orfor
Person
alArea
only)
(i)Lo
wband
width
(2Mbp
s)(ii)U
sedin
medicaldevicesa
ndindu
strialsensorsLo
wpo
wergood
forw
earables
(iii)Usablefor
Realtim
elocationsyste
msw
ithmedium
accuracy
Zigbee
Yes(30-300
feet)
No(orfor
Person
alArea
only)
(i)Lo
wdatarate
(ii)Ind
ustrialand
someh
omea
pplications
(egho
mee
nergy
mon
itorin
gwire
lesslig
htsw
itches)
(iii)Lo
wtransm
itpo
werLo
wbatte
ryconsum
ption
NoteAfewotherlegacyIoTwire
lesstechno
logies
exist
(egCat0Cat1EC
-GSM
Weightless)b
utaren
otinclu
dedin
thistable
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 9
Table2Con
tinued
Techno
logy
Indo
orusability
Outdo
orusability
Basic
aspects
Wi-F
iYes300feet
Tosomed
egreerequ
ires
inter-spot
conn
ectiv
itybackbo
ne(w
iredor
wire
less)(eg
80211ah
dista
ncer
ange
upto
abou
t12
mile)
(i)Severalbands
(ii)In2018
theF
CCallowe
dthee
xpansio
nof
the6
GHzb
andto
next-generationWi-F
idevices
with
12GHzo
fadd
ition
alspectrum
spanning
5925to
7125
GHz(
currentW
i-Fin
etwo
rkso
perateat24
GHza
nd5GHzw
ithafew
vend
orso
fferin
g60
GHzldquo
WiGigrdquothis
having
arange
of30
feetndashIEEE
80211a
dandIEEE
80211a
y)(iii)Highadop
tion
most(bu
tnot
all)indo
orIoTutilize
Wi-F
igood
functio
nality
(iv)F
reeldquo
airtim
erdquo(v)S
ubjectto
interfe
rencemalicious
orno
n-malicious
interfe
rence
(egtoo
manyho
tspo
ts)couldim
pairthes
ensorfrom
send
ingdata
either
onafi
ne-grain
orcoarse-grain
basis
Bluetooth
Yes30
feet
No(orfor
Person
alArea
only)
(i)Lo
wband
width
(2Mbp
s)(ii)U
sedin
medicaldevicesa
ndindu
strialsensorsLo
wpo
wergood
forw
earables
(iii)Usablefor
Realtim
elocationsyste
msw
ithmedium
accuracy
Zigbee
Yes(30-300
feet)
No(orfor
Person
alArea
only)
(i)Lo
wdatarate
(ii)Ind
ustrialand
someh
omea
pplications
(egho
mee
nergy
mon
itorin
gwire
lesslig
htsw
itches)
(iii)Lo
wtransm
itpo
werLo
wbatte
ryconsum
ption
NoteAfewotherlegacyIoTwire
lesstechno
logies
exist
(egCat0Cat1EC
-GSM
Weightless)b
utaren
otinclu
dedin
thistable
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
10 Wireless Communications and Mobile Computing
MCO
Analytics
LoRaSigfox
NB-IoTLTE-M
IoT
LoRaSigfox NB-IoT
LTE-M
IoT
IoTIoT
IoT
IoT
IoTIoT
5G
5G
5G
5G
5G IoT
Backhaul
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
Distributed City-wide In-building services
5G IoT
5G IoT
5G IoT
5G IoT
5G IoT
IoT
5G IoT
5G IoT
DAS
Wi-Fi
DAS
DASIoT
IoT
IoT
IoT
IoT
Figure 2 The pre-5G and the 5G IoT connectivity ecosystem
4GLTE and 5G are expected to coexist for many yearsHowever it is fair to say that like many other technologiesbefore 5G this technology is probably going through a ldquohype-cyclerdquo where a technology is supposed to be ldquoall things toall peoplerdquo and be the ldquobe-all-and-end-all technologyrdquo bothclaims will be abrogated in time Proponents argue that 5Gwill ldquomaximize the satisfaction of end-users by providingimmersiveness intelligence omnipresence and autonomyrdquo
21 5G Standardization and Use Cases Standardization workfor 5G systems has been undertaken by several internationalbodies with the goal of achieving one unified global standardMany well-known research centers universities standardsbodies carriers and technology providers have been involvedin advancing the development of the technology for a2020 rollout including the Internet Engineering Task Force(IETF) the Open Network Automation Platform (ONAP)theGSMA and the EuropeanTelecommunications StandardsInstitute Network Function Virtualization (ETSI NFV) Inparticular work on 5G requirements services and technicalspecifications has been undertaken in the past few yearsby three key entities (i) International TelecommunicationUnion-Radio Communication Sector (ITU-R) [30] (ii) NextGeneration Mobile Networks (NGMN) Alliance [31] and(iii) the 3rd Generation Partnership Project (3GPP) [32]TheITU-R has assessed usage scenarios in three classes ultra-reliable and low-latency communications (URLLC) mas-sive machine-type communications (mMTC) and enhancedmobile broadband (eMBB) eMBB is probably the earliest
class of services being broadly supported and implementedKey performance indicators are identified for each of theseclasses such as spectrum efficiency area traffic capacityconnection density user-experienced data rate peak datarate and latency among others The ability to efficientlyhandle device mobility is also critical Some examples ofeMBB use cases include Non-SIM devices smart phoneshomeenterprisevenues applications UHD (4K and 8K)broadcast and virtual realityaugmented reality mMTCuse cases include smart buildings logistics tracking fleetmanagement and smart meters URLLC cases include trafficsafety and control remote surgery and industrial control 5Gsystems are expected to support
(i) Tight latency availability and reliability requirementsto facilitate applications related to video deliveryhealthcare surveillance and physical security logis-tics automotive locomotion and mission-criticalcontrol among others particularly in an IoT context
(ii) A panoply of data rates up tomultiple Gbps and tensof Mbps to facilitate existing and evolving applica-tions particularly in an IoT context
(iii) Network scalability and cost-effectiveness to supportboth clustered users with very high data rate require-ments as well a large number of distributed deviceswith low complexity and limited power resourcesparticularly in an IoT context where as noted arapid increase in the number of connected devices isanticipated and
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 11
Table 3 Radio interface goals as defined in IMT-2020
(i) MR for downlink peak data rate is 20 Gbps(ii) MR for uplink peak data rate is 10 Gbps(iii) Target downlink ldquouser experienced data raterdquo is 100 Mbps(iv) Target uplink ldquouser experienced data raterdquo is 50 Mbps(v) Downlink peak spectral efficiency is 30 bpsHz(vi) Uplink peak spectral efficiency is 15 bpsHz(vii) MR for user plane latency for eMBB is 4ms(viii) MR for user plane latency for URLLC is 1ms(ix) MR for control plane latency is 20ms (a lower control plane latency of around 10ms is encouraged)(x) Minimum requirement for connection density is 1000000 devices per km2
(xi) Requirement for bandwidth is at least 100 MHz(xii) Bandwidths up to 1 GHz are required for higher frequencies (above 6 GHz)MR = Minimal RequirementSource ITU-R SG05 Contribution 40 ldquoMinimum requirements related to technical performance for IMT-2020 radio interface(s)rdquo Feb 2017
(iv) Pragmatic deployment cost metrics along with ac-ceptable service price points across the gamut ofapplications and data rates particularly in an IoTcontext
Specifically some of the design details are a latency below5 msec (as low as 1 msec) support for device densities ofup to 100 devicesm2 reliable coverage area integration oftelecommunications services including mobile fixed opti-cal and MEOGEO satellite and seamless support for theIoT ecosystem For example the technical objective 5G asenvisioned ofMETIS (Mobile andWireless CommunicationsEnablers for the Twenty-twenty Information Society -- aEuropean Community advocacy effort related to mobility)are as follows [47ndash54]
(i) 1000 x higher mobile data volume per area than cur-rent systems
(ii) 10 to 100 x higher number of devices than currentsystems (ie dense coverage)
(iii) 10 to 100 x higher user data rate than current systems(eg 1-20 Gbps)
(iv) 10 x longer battery life for low power IoT devicesthan current systems (up to a 10-year battery life formachine type communications) and
(v) 5 x reduced end-to-end latency than current systems
Table 3 defines the 5G radio interface goals as defined in IMT-2020 A number of these requirements are in fact being met(in various measure) by the systems now being deployedTheexpectation is that to provide the full panoply of 5G servicessignificant changes in both wireless technologies and corenetworks will be required
As a point of observation 3GPPTR 22891 has definedandor described the following service groups eMBB Crit-ical Communication mMTC Network Operations andEnhancement of Vehicle-to-Everything (V2X) NGMN hasdefined andor described the following service groupsBroadband access in dense area Indoor ultra-high broad-band access Broadband access in a crowd 50+ Mbps every-where Ultra low-cost broadband access for low ARPU areas
Mobile broadband in vehicles Airplanes connectivity Mas-sive low-cost Low long-rangelow-power MTC BroadbandMTC Ultra low latency Resilience and traffic surge Ultra-high reliability and Ultra low latency Ultra-high availabilityand reliability and Broadcast-like services
Figure 3 depicts some of the key 5G services that can beutilized for the IoT in themedium term in Smart Cities otherservices shown might also be used over time Although somehave associated Smart Cities with mMTC we are of the opin-ion that the early applications will be more within the eMBBdomain (some others also agree [55]) Also one would expecteMBB to be deployedmore broadly driven by the commercialldquoappealrdquo of the video services it facilitates Augmented andorvirtual reality (ARVR) are emerging as keys application of5G networks also involving some IoT aspects To meet therequirements of lower latency and massive data transmissionin ARVR applications software-defined networking (SDN)with a multi-path cooperative route (MCR) scheme thatminimizes delay may be ideally positioned for 5G small cellnetworks [56] Note parenthetically that video requirementsrange from about 8 Mbps for HD 25 Mbps for UHD50 Mbps for 360-degree UHD video 200 Mbps for 360-degree HDR (high dynamic range) video and up to 1 Gbpsfor 6DoFMPEG-I The evolving MPEG-I Visual standardaddresses visual technologies of immersive media 360 videoprovides panoramic video texture projected onto a virtualshape surrounding the userrsquos head from which the uservisualizes a portion for an immersive video experience 6DoF(6 Degrees of Freedom) supports movements along threerotation axes and three translations and presumes that fullfreedom of movement through the scene is possible [57]5GeMBB may eventually support some (but not necessarilyall) of these video applications but these applications are wellbeyond the IoT applications discussed in this paper IP-basedvideo surveillance in Smart Cities that may be supported byIoT can operate rather well at the 0384-25 Mbps bandwidthrange
Figure 4 highlights some technical features of 5G servicesthat can be utilized for the IoT in Smart Cities in terms ofdata rates latency reliability device density and so on 5G IoTovercomes the well-known limitation of unlicensed LPWAN
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
12 Wireless Communications and Mobile Computing
NGMNITU-R M2083
3GPP
TR 2
289
1
High likelihood ofIoT usage inSmart Cities
in the short term
Medium likelihood ofIoT usage inSmart Cities
in the short term
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-
Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbps everywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
Figure 3 5G services that can be utilized for the IoT in Smart Cities
technologies that utilize crowded license-free frequencybands especially in large cities therefore 5G IoT is ideal forSmart City for mission-critical and Quality of Service (QoS)-aware applications (for example traffic management smartgrid utility control)
22 5G Evolution 3GPP has specified new 5G radio accesstechnology 5G enhancements of 4G (fourth generation)networks and new 5G core networks Specifically it hasdefined a new 5GCore network (5GC) and a new radio accesstechnology called 5G ldquoNewRadiordquo (NR)Thenew 5GC archi-tecture has several new capabilities built inherently into itas native capabilities multi-Gbps support ultra-low latencyNetwork Slicing Control and User Plane Separation (CUPS)and virtualization To deploy the 5GC new infrastructurewill be needed There is a firm goal to support for ldquoforwardcompatibilityrdquo The 5G NR modulation technique and framestructure are designed to be compatible with LTEThe 5GNRduplex frequency configuration will allow 5G NR NB-IoTand LTE-M subcarrier grids to be aligned This will enablethe 5G NR user equipment (UE) to coexist with NB-IoT andLTE-M signals As might be expected however it is possibleto integrate into 5G elements of different generations anddifferent access technologiesndash two modes are allowed the SA(standalone) configuration and the NSA (non-standalone)configuration (see Figure 5 also positioning IoT support)
(i) 5G Standalone (SA) Solution in 5G SA an all new 5Gpacket core is introduced SA scenarios utilize onlyone radio access technology (5G NR or the evolved
LTE radio cells) the core networks are operatedindependently
(ii) 5G Non-Standalone Solution (NSA) in 5G NSAOperators can leverage their existing Evolved PacketCore (EPC)LTE packet core to anchor the 5G NRusing 3GPP Release 12 Dual Connectivity featureThis will enable operators to launch 5G more quicklyand at a lower cost This solution might sufficefor some initial use cases However 5G NSA hasa number of limitations thus these Operators willeventually be expected to migrate to 5G Standalonesolution NSA scenario combines NR radio cells andLTE radio cells using dual-connectivity to provideradio access and the core network may be either EPCor 5GC
Multiple evolutiondeployment paths may be employed byservice providers (service providers of various servicesincluding IoT services) to reach the final target configu-ration this migration could well take a decade and mayalso have different timetables in various parts of a countryeg top urban areas top suburban areas secondary urbanareas secondary suburban areas exurbian areas rural areasFigure 6 depicts the well-known migration paths The IoTimplementerwill need to be keenly aware of what 5G (5G IoT)services are available in a given area as an IoT implementationis contemplated In Figure 6 Scenario 1 illustrates that theIoT Service provider will continue to use LTE and EPC toprovide services (eg NB-IoT) here only legacy IoT devicescan be supported The provider only has a standalone radio
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 13
NGMNITU-R M2083
3GPP
TR 2
289
1
Enhanced MobileBroadband
MassiveMachine-type
Communications (MTC)
Ultra-reliable andLow Latency
Communications
EnhancedMobile
Broadband
CriticalCommunications
MassiveMachine-type
Communications
NetworkOperations
Enhancementof Vehicle-to-Everything
Broadband accessIn dense area
Indoor ultra-highbroadband accessBroadband access
in a crowd
Ultra-low-cost broadbandin low ARPU areas
50+ Mbpseverywhere
Resilience and traffic surge
Mobile broadband invehicles
Airplane ConnectivityMassive low-costlong-rangelow power MTC
Broadband MTC
Ultra low latency
Broadcast-line services
Ultra high reliability andultra low latency
Ultra high availability andreliability
LatencyData Rate Traffic Density ConnectionDensity
Mobility
Very lowVery High(eg peak
rate 10 GbpsHigh
High (eg
simultaneously500 kmh
User ExperiencedData Rate
DataRate
Area TrafficCapacity
ConnectionDensityMobility
HighHigh High Medium
SpectrumEfficiency
High
Latency
Medium
Network EnergyEfficiency
High High
User ExperiencedData Rate
TrafficDensity
ConnectionDensityMobility
DL 300 MbpsUL 50 Mbps
100 kmh(Activity factor 10)
End-to-endLatency
10 ms
DL 1 GbpsUL 500 Mbps
Pedestrian(7 kmh) (Activity factor 30)10 ms
ReliabilityLatency Traffic Density PositionAccuracy
Ultra highLow
(eg 1 msend-to-end
Precise positionwithin 10 cm
High (eg10000
2500kG2
75000kG2
DL 750 GbpskG2
UL 125 GbpskG2
DL 15 TbpskG2
UL 2 TbpskG2
2500kG2 50
sensors 10 kG2
Figure 4 Some technical features of 5G services that can be utilized for the IoT in Smart Cities
CoreNetwork
RadioAccessNetwork
5GC
EPC
SA
NSA
Newcore
transport
Legacy core
transport
NewIoT
access
LegacyIoT
access
Core
3GPP has defined a new 5G core network (5GC) and a new radio accessTechnology known as 5G ldquoNew Radiordquo (NR)
Access
5G Standalone (SA) solution In 5G SA an all new 5G packet core is introducedSA scenarios utilize only one radio access technology (5G NR or the evolved LTEradio cells) the core networks are operated independently
5G Non-Standalone Solution (NSA) in 5G NSA Operators can leverage theirexisting Evolved Packet Core (EPC)LTE packet core to anchor the 5G NR using3GPP Release 12 Dual Connectivity feature
Figure 5 5G Transition Options and IoT support
technology in this case LTE only Scenario 2 illustrates an IoTService provider has migrated completely to NR (again onlyproviding a standalone radio technology) but will retain theexisting core network the EPC (Only) new 5G IoT devicescan be used In scenarios 5 and 6 the service providers willsupport both the legacy LTE and the new NR (clearly inthis non-standalone arrangement both radio technologiesare deployed) Some of these providers retain the legacy coreand some will deploy the new 5GC core Both legacy and 5GIoT devices can be supported
3GPP approved the 5G NSA standard at the end of 2017and the 5G SA standard in early 2018 in the context ofits Release 15 Release 15 also included the support eMBBURLLC and mMTC in a single network to facilitate thedeployment of IoT services Release 15 also supports 28 GHzmillimeter-wave (mmWave) spectrum and multi-antennatechnologies for access
23 5G Frequency Bands Focusing on the radio technologythere are number of spectrum bands that can be used in
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
14 Wireless Communications and Mobile Computing
Legacy IoTdevice (4G)
New IoTdevice (5G)
Legacy IoTdevice (4G)
New IoTdevice (5G)
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s1
s2
s3
s4SA
LTE
NR
EPC
5GC
Cloud
IoT analytics
Core
Access
s5
s6NSA
amp
Figure 6 Detailed 5G Transition Options and IoT support
5G these bands can be grouped into three macro categoriessub-1 GHz 1-6 GHz and above 6 GHz The more advancedfeatures especially higher data rates require the use ofthe millimeter wave spectrum New mobile generations aretypically assigned new frequency bands and wider spectralbandwidth per frequency channel (1G up to 30 kHz 2Gup to 200 kHz 3G up to 5 MHz and 4G up to 20 MHz)Up to now cellular networks have used frequencies below6 GHz Generally without advanced MIMO (Multiple InMultiple Out) antenna technologies one can obtain about10 bits-per-Hertz-of-channel bandwidth But the integrationof new radio concepts such as Massive MIMO Ultra DenseNetworks Device-to-Device and mMTC will allow 5G tosupport the expected increase in the data volume in mobileenvironments and facilitate new IoT applications Imple-mentable standards for 5G are being incorporated in 3GPPRelease 15 onwards As noted 3GPP Rel 15 defines New 5GRadio and Packet Core evolution to facilitate interoperabledeployment of the technology
The millimeter wave spectrum also known as ExtremelyHigh Frequency (EHF) or more colloquially mmWave isthe band of electromagnetic spectrum running between 30GHz and 300 GHz Bands within this spectrum are beingconsidered by the ITU and the Federal CommunicationsCommission in the US as a mechanism to facilitate 5G bysupporting higher bandwidthThe use of a 35 GHz frequencyto support 5G networks is also gaining some popularitybut he higher speeds networks will use other frequencybands including millimeter-wave frequencies (these bandsranging from 28 GHz to 73 GHz specifically the 28 3739 60 and 72ndash73 GHz bands) In the US recently theFCC approved spectrum for 5G including millimeter-wavefrequencies in the 28 GHz 37 GHz and 39 GHz bandsalthough these targeted cellular frequencies may nominally
overlap with other pre-existing users of the spectrum forexample point-to-point microwave paths Direct Broadcastsatellite TV and high throughput satellite (HTS) systems (Ka-band transmissions)
Initially 5G will in many cases use the 28 GHz bandbut higher bands will very likely be utilized later on ini-tial implementations will support a maximum speed of 1Gbps Lower frequencies (at the so-called C band) are lesssubject to weather impairments can travel longer distancesand penetrate building walls more easily Waves at higherfrequencies (Ku Ka and EV bands) do not naturally travel asfar or penetrate walls or objects as easily However a lot morechannel bandwidth is available in millimeter-wave bandsFurthermore developers see the need for ldquoan innovativeutilization of spectrumrdquo ldquosmall cellrdquo approaches are requiredto address the scarcity of the spectrum but at the same timecovering the geography V band spectrum covers 57-71 GHzwhich in many countries is an ldquounlicensedrdquo band and E bandspectrum covers 71-76 GHz 81-86 GHz and 92-95 GHz
In the US in 2018 the FCC also opened up as anldquointerimrdquo step for 5G a ldquomid-bandrdquo radio spectrum at35 GHz which was previously reserved for naval radaruse The 35 GHz band provides a combination of signalpropagation distance acceptable building penetration andincreased bandwidth The FCC created 15 channels withinthe 3550-3700 GHz band auctioning seven channels toldquopriority access licensesrdquo andmaking eight channels availablefor general access -- the US Navy still getting priority acrossthe band when and as needed With this approval 5G devicescan be built to support the same 35GHz ranges across NorthAmerica Europe and Asia [58]
In addition to new bands 5G technology is expected touse beam-forming and beam-tracking where a cellrsquos antennacan focus its signal to reach a specific mobile device and
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 15
10 km
1 km
01 km
90
100
110
120
130
140
150
160
170Pa
th L
oss (
dB)
102101
Frequency (GHz)
Figure 7 Path loss as a function of distance and frequency
then track that device as it moves Beamforming utilizesa large number (hundreds) of antennas at a base stationto achieve highly directional antenna beams that can beldquosteeredrdquo in a desired direction to optimize transmissionand throughput performance Massive MIMO is a systemwhere a transmission node (base station) is equipped witha large number (hundreds) of antennas that simultaneouslyserve multiple users with this technology multiple messagesfor several terminals can be transmitted on the same time-frequency resource
24 5G Transmission Characteristics at Higher FrequenciesDue to RF propagation phenomena that are more pro-nounced at the higher frequencies such as multipath prop-agation due to outdoor and indoor obstacles free spacepath loss atmospheric attenuation due to rain fog and aircomposition (eg oxygen) small cells will almost invariablybe needed in 5G environments especially in dense urbanenvironments Additionally Line of Sight (LOS) will typicallybe required ITU-R P series of recommendations has usefulinformation on radio wave propagation including ITU-RP838-3 2005 ITU-R P840-3 2013 ITU-R P676-10 2013and ITU-R P525-2 1994 Figures 7 8 9 and 10 highlight theissues at the higher frequencies including the millimeter-wave frequencies Figure 7 depicts the path loss as a functionof distance and frequency Figure 8 shows the attenuation asa function of precipitation and frequency Figure 9 illustratesthe attenuation as a function of fog density and frequencyFigure 10 depicts the attenuation as a function of atmosphericgases and frequency (notice high attenuation around 60GHz)
In addition to the broad service requirements brieflyhighlighted in Table 3 (for example latency user densitydistribution etc) there are specific IoT nodal considerationsthat have to be taken into account as one develops the nextgeneration network For example IoT nodes typically arelow-complexity devices and have limited on-board power5G systems have to take these restrictions and considerations
Extreme Rain
Heavy Rain
Moderate Rain
Light Rain
101 102
Frequency (GHz)
10minus2
10minus1
100
101
102
Rain
Atte
nuat
ion
(dB
km)
Figure 8 Attenuation a function of precipitation and frequency
Heavy
Medium
10minus3
10minus2
10minus1
100
101
Fog
Atte
nuat
ion
(dB
km)
101 102
Frequency (GHz)
Figure 9 Attenuation a function of fog density and frequency
into account Table 4 provides a summary of some of theseconsiderations and the 5G support
3 Small Cell and Building Penetration Issues
As expected communications at mmWave frequencies haveattracted a lot of interest due to the large available spectrumbandwidth that can potentially result in multiple gigabit persecond transmissions per user This follows a similar trend
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
16 Wireless Communications and Mobile Computing
Atm
osph
eric
Gas
10minus2
10minus1
100
101
102
Atte
nuat
ion
(dB
km)
101 102Frequency (GHz)
Figure 10Attenuation a function of atmospheric gases and frequency(notice high attenuation around 60 GHz)
in satellite communications with the introduction of Ka ser-vices especially HTSs High bandwidth will typically requirea wide spectrum Millimeter wave frequencies (signals withwavelength ranging from 1 millimeter to 10 millimeters) sup-port a wide usable spectrum The millimeter wave spectrumincludes licensed lightly licensed and unlicensed portionsBandwidth demand and goals are the main driver for theneed to use the millimeter wave spectrum particularly foreMBB-based applications allowing users to receive 100Mbpsas a bare minimum and 20 Gbps as a theoretical maximumThe use of millimeter wave frequencies however will implythe use of a much smaller tessellation of cells and supportivetowers or rooftop transmitters due as noted to transmissioncharacteristics such as high attenuation and directionalityThis is an important design consideration for 5G especiallyin dense cityurban environments The aggregation of thesetowers will by itself require a significant backbone networkwhether a mesh based on some point-to-point microwavelinks an fiber network or a set of ldquowireless fiberrdquo linksMillimeter wave system utilize smaller antennas comparedto systems operating at lower frequencies the higher fre-quencies in conjunction withMIMO techniques can achievesensible antenna size and cost The millimeter wave tech-nology can be utilized both for indoors and outdoors high-capacity fixed or mobile communication applications Theterm ldquodensificationrdquo is also used to describe the massivedeployment of small cells in the near future
MmWave products used for backhauling typically operateat 60 GHz (V Band) and 7080 GHz (E Band) and offer solu-tions in both Point to Point and Point to Multipoint (PtMP)configurations providing end to end multi-gigabit wirelessnetworks for example 1 Gbps up to 10 Gbps symmetric per-formance Very small directional antennas typically less thana half-square foot in area are used to transmit andor receive
signals which are highly focused beams stationary radiosystems are often installed on rooftops or towers MmWaveproducts are now appearing on the market targeting highcapacity Smart City applications 5G Fixed Gigabit WirelessAccess solutions and Business Broadband Urban canyonshowever may limit the utility of this technology to very shortLOS paths Mobile applications of mmWave technology aremore challenging On the other hand one advantage of thistechnology is that short transmission paths (high propagationlosses) and high directionality allow for spectrum reuse bylimiting the amount of interference between transmittersandor adjacent cells Near LOS (NLOS) applications may bepossible in some cases (especially for short distances)
Currently mm wave frequencies are being utilized forhigh-bandwidth indoor applications for example streaming(ldquomiracastingrdquo) of HD or UHD video and VR support(eg using 80211ad Wi-Fi) Traditionally these frequencieshave not been used for outdoor broadband applicationsdue to high propagation loss multipath interference andatmospheric absorption (gases rain fog and humidity) citedabove in addition the practical transmission range is a fewkilometers in open space [68] Recently the FCC proposednew rules for wireless broadband in wireless frequenciesabove 24 GHz stating that it is ldquotaking steps to unlock themobile broadband and unlicensed potential of spectrum at thefrontier above 24 GHzrdquo [69] The ITU and the 3GPP havedefined two-phases of research the first phase (expected tocomplete by press time) is to assess frequencies less than40 GHz to address short-term commercial requirements thesecond phase entails assessing the IMT 2020 requirements bystudying frequencies up to 100 GHzThe following mmWavebands being considered among other bands [70]
(i) 7 GHz of spectrum in total in the band 57 GHz to 64GHz unlicensed
(ii) 34 GHz of spectrum in total in the 28 GHz38 GHzlicensed but underutilized region
(iii) 129 GHz of spectrum in total 71 GHz81 GHz92 GHzlight-licensed band
Following the most recent World RadiocommunicationsConference the ITU also identified a list of proposedglobally-usable frequencies between 24 GHz and 86 GHzas follows 2425ndash275 GHz 318ndash334 GHz 37ndash405 GHz405ndash425 GHz 455ndash502 GHz 504ndash526 GHz 66ndash76 GHzand 81ndash86 GHz
31 Cell Types MmWave transmission will drive the require-ment for small cells [71 72] ldquoSmall cellsrdquo refer to relativelylow-powered radio communications equipment (base sta-tions) and ancillary antennas andor towers that providemobile internet and IoT services within localized areasSmall cells typically have a range up to one-to-two kilometersbut can also be smaller -- on the other hand a typical mobilemacrocell (such as urban macro-cellular [UMa] or ruralmacrocell [RMa]) has a range of several kilometers up to 10-to-20 of kilometers) The terms femtocells picocells micro-cells urban microcell (UMi) and metrocells are effectivelysynonymous with the ldquosmall cellsrdquo concept Small(er) cells
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 17
Table 4 Example of IoT nodal considerations for 5G systems
IoT device issue 5G Support
Low complexity devices Broad standardization leads to simplification eg SOC (System on a Chip)andor ASIC (Application Specific IC) development
Limited on-board power Technology allows a battery life sim10 yearsDevice mobility Good mobility support in a cellular5G systemOpen environment Broad standardization leads to broad acceptance of the technology
Devices universe by type and bycardinality
Standardized air interfaces can reduce certain aspects of the end-node justlike Ethernet simplified connectivity to a network regardless of thefunctionality of the processor per se
Always connectedalways on mode ofoperation Cost-effective connectivity services allow the always on mode of operation
IoT security (IoTSec) concerns [59 60]
Security capabilities are being added The use of 256-bit symmetriccryptography mechanisms is expected to be fully incorporatedTheencryption algorithms are based on SNOW 3G AES-CTR and ZUC andintegrity algorithms are based on SNOW 3G AES-CMAC and ZUCThemain key derivation function is based on HMAC-SHA-256 Identitymanagement (eg via the 5G authentication and key agreement [5G AKA]protocol andor the Extensible Authentication Protocol [EAP]) Privacy(conforming to the General Data Protection Regulation [GDPR]) andSecurity assurance (eg using Network Equipment Security AssuranceScheme [NESAS]) are supported Some of these mechanisms are described[61ndash65] As another example the ETSI Technical Committee onCybersecurity issued in 2018 two encryption specifications for accesscontrol in highly distributed systems such as G and IoT Attribute-BasedEncryption (ABE) that describes how to secure personal data
Lack of agreed-upon end-to-endstandards
Broad standardization possible with 5G if the technology is broadlydeployed and is cost-effective
Lack of agreed-upon end-to-endarchitecture
Standardization at the lower layers (Data Link Control and Physical) candrive the development of a more inclusive multi-layer multi-applicationarchitecture
have been used for years to increase area spectral efficiency-- the reduced number of users per cell provides more usablespectrum to each user However the smaller cells in 5G arealso dictated by the propagation characteristics In the 5Gcontext UMi typically have radii of 5-120 meters for LOSand 20 to 270 meters in NLOS UMa typically have radiiof 60-1000 meters for LOS and 50-1500 meters for NLOS[73] Given their size 5GmmWave UMi cells will be able tosupport high bandwidth enabling eMBB services over smallareas of high traffic demand At themmWave operation user-device proximity with the antenna will enable higher signalquality lower latency and by definition high data rates andthroughput Also to be notedmmWave frequenciesmake thesize of multi-element antenna arrays practical enabling largeMulti-user MIMO (MU-MIMO) solutions
Signal penetration indoors may represent a challengejust as is the case even at present with 3G4G LTE even fortraditional voice and internet access and data services Thishas driven the need for DAS systems especially in densely-constructed downtown districts Free space attenuation atthe higher frequency power budgets directionality require-ments and weather all impact 5G and 5G IoT Outdoor smallcells and building-resident Distributed Antenna Systems(DAS) systems utilize high-speed fiber optic lines or ldquowirelessfiberrdquo to interconnect the sites to the backbone and theInternet cloud
Figure 11 depicts a 5G IoT ecosystem where mmWavetechnology is used Figure 12 shows typical (4G LTE) urbanmicrocell towers Figure 13 depicts a Smart City supported via(5G) urban microcells
32 Assessment of Transmission Issues Reference [74] pro-vides a fairly comprehensive assessment of the transmissionchannel issues as they apply to 5G The importance of thistopic is accentuated by the large number of agencies activelyresearching this topic including [55 73ndash87]
(i) METIS(ii) 3GPPP(iii) MiWEBA (Millimetre-Wave Evolution for Backhaul
and Access)(iv) ITU-R M(v) COST2100(vi) IEEE 80211(vii) NYU WIRELESS interdisciplinary academic re-
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
18 Wireless Communications and Mobile Computing
4GLTE
mmWavebackhaul
Other
backhauls
5G IoT
mmWavebackhaul
5G mmWave5G IoT5G IoT
5G mmWave
5G mmWave
5G mmWave
5G lsquomidbandrsquo
MCO
mmWavebackhaul
Figure 11 The 5G IoT ecosystems
Microcell towers usable in 5G and 5G IoT
Figure 12Microcell towers (these for 4G but a lotmore for 5G) (non-copyrighted material from FCC-related filings [91])
(xi) 5G mmWave Channel Model Alliance (NIST initi-ated North America based)
(xii) mmMAGIC (Millimetre-Wave Based Mobile RadioAccess Network for Fifth Generation IntegratedCommunications) (Europe based)
(xiii) IMT-2020 5G promotion association (China based)
(also including firms and academic centers such as but notlimited to ATampT Nokia Ericsson Huawei IntelFraunhofer
Figure 13 Microcells for 5G5G IoT
HHINTTDOCOMOQualcommCATT ETRI ITRICCUZTE Aalto University and CMCC)
Diffraction loss (DL) and frequency drop (FD) are justtwo of the path quality issues to be addressed Althoughgreater gain antennas will likely be used to overcome pathloss diffuse scattering from various surfaces may introducelarge signal variations over travel distances of just a fewcentimeters with fade depths of up to 20 dB as a receivermoved by a few centimeters These large variations of thechannel must be taken into consideration for reliable design
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 19
Distance Between Transmitter and Receiver (m)500010 30 50 100 200 500 1000
Path Loss results as obtained by5GCM 3GPP METIS simulationsunder various conditions at 28 GHzfall between these two boundary lines
150
70
90
110
130
150
170
Path
Los
s (dB
)
Figure 14 Path Loss simulations for 5G by various entities
of channel performance including beam-formingtrackingalgorithms link adaptation schemes and state feedback algo-rithms Furthermore multipath interference from coincidentsignals can give rise to critical small-scale variations in thechannel frequency response In particular wave reflectionfrom rough surfaces will cause high depolarization ForLOS environment Rician fading of multipath componentsexponential decaying trends and quick decorrelation in therange of 25 wavelengths have been demonstrated Further-more received power of wideband mmWave signals has astationary value for slight receiver movements but averagepower can change by 25 dB as the mobile transitions arounda building corner from NLOS to LOS in an UMi settingAdditionally human body blockage causes more than 40 dBof fading at the mmWave frequencies Figure 14 depicts thepath loss according to various simulations for 5G by variousstakeholder entities
Themain parameter of the radio propagationmodel is thePath Loss Exponent (PLE) which is an attenuation exponentfor the received signal PLE has a significant impact on thequality of the transmission links In the far field region ofthe transmitter if PL(d0) is the path loss measured in dB at adistance d0 from the transmitter then the loss in signal powerexpected when moving from distance d0 to d (dgtd0) is [88ndash90] is
1198751198711198890997888rarr119889 (119889119861) = 119875119871 (1198890) + 10119899 log10 ( 1198891198890) + 120594119889119891 le 1198890 le 119889
(1)
where
PL(d0) = Path Loss in dB at a distance d0n = PLE120594 = A zero-mean Gaussian distributed random vari-able with standard deviation 120590 (This is utilized onlywhen there is a shadowing effect if there is noshadowing effect then this random variable is takento be zero)
See Figure 15 Usually PLE is considered to be known upfrontbut in most instances PLE needs to be assessed for the caseat hand It is advisable to estimate the PLE as accuratelyas possible for the given environment PLE estimation isachieved by comparing the observed values over a sampleof measurements to the theoretical values Obstacles absorbsignals thus treating the PLE as a constant is not an accuraterepresentation of the real environments both indoors andoutdoors (for example treating PLE as a constant whichmay cause serious positioning errors in complicated indoorenvironments [88]) Usually to model real environments theshadowing effects cannot be overlooked by taking the PLEas a constant (a straight-line slope) To capture a shadowingeffect a zero-mean Gaussian random variable with standarddeviation 120590 is added to the equation Here the PLE (slope)and the standard deviation of the random variable should beknown precisely for a better modeling
Table 5 provides theoretical performance equationsdeveloped by 3GPP and ETSI for outdoor channel perfor-mance [81] As pragmatic working parameters one has thefollowing
(i) PLE values are in the 19 and 22 range for LOS and atthe 28 GHz and 60 GHz bands PLE is approximately45 and 42 range forNLOS in the 28GHz and 60GHzbands
(ii) Rain attenuation of 2-20 dBkm can be anticipated forrain events ranging from light rain (125 mmhr) todownpours (50mmhr) at 60GHz (higher for tropicalevents) For 200-meter cells the attenuation will bearound 02 db for 5mmhr rain at 28 GHz and 09 dBfor 25mmhr rain at 28 GHz The attenuation will bearound 05 db for 5mmhr rain at 60 GHz and 2 dBfor 25mmhr rain at 60 GHz
(iii) Atmospheric absorption of 1-10 dBkm occurs atthe mmWave frequencies For 200-meter cells theabsorption will be 004 dB at 28 GHz and 32 dB at60 GHz
20 Wireless Communications and Mobile Computing
Table 5 Path Loss Equations for mmWave 5G5G IoT
ℎBS
d3D-out
d2D-out
d3D-in
d2D-in
ℎUT
Scenario LOSNLOS Pathloss [dB] (119891119888 is in GHz and 119889 is in meters) Shadow fadingstd [dB]
Applicability rangeantenna heightdefault values
UMi - Street Canyon LOS
119875119871UMi-LOS =1198751198711 10m le 1198892D le 1198891015840BP1198751198712 1198891015840BP le 1198892D le 5km
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
InH - OfficeLOS 119875119871 InH-LOS = 324 + 173 log10 (1198893D) + 20 log10 (119891119888) 120590SF = 3 1m le 1198893D le 100m
NLOS
119875119871 InH-NLOS = max (119875119871 InH-LOS 1198751198711015840InH-NLOS)1198751198711015840InH-NLOS =383 log10 (1198893D) + 1730 + 249 log10 (119891119888)120590SF = 803 1m le 1198893D le 86m
Optional1198751198711015840InH-NLOS = 324 + 20 log10(119891119888) + 319 log10(1198893D) 120590SF = 829 1m le 1198893D le 86m
Note 1 Breakpoint distance 1198891015840BP = 4ℎ1015840BSℎ1015840UT119891119888119888 where 119891119888 is the centre frequency in Hz 119888 = 30 times 108 ms is the propagation velocity in free
space and ℎ1015840BS and ℎ1015840UT are the effective antenna heights at the BS and the UT respectively The effective antenna heights ℎ1015840BS and ℎ1015840UT are computedas follows ℎ1015840BS = ℎBS minus ℎE ℎ
1015840UT = ℎUT minus ℎE where ℎBS and ℎUT are the actual antenna heights and hE is the effective environment height For
UMi ℎE = 10m For Uma ℎE = 1m with a probability equal to 1(1 + C(1198892D ℎUT)) and chosen from a discrete uniform distribution uniform(12 15 (ℎUT-15)) otherwise With C(1198892D ℎUT) given by 119862(1198892D ℎUT) = 0 ℎUT lt 13m ((ℎUT minus 13)10)
15119892(1198892D) 13m le ℎUT le 23m where119892(1198892D) = 0 1198892D le 18m (54)(1198892D100)
Free Space PLE 20Uma cell PLE 27 ndash35Indoor LOS PLE 17 ndash18Indoor obstructed PLE 4 ndash6
0L0N
(dB)
FIA10 (>)
- 10 n FIA10(>)
Figure 15 PLE
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 21
Penetration into buildings is an issue for mmWave commu-nication this being a lesser concern for contemporary sub 1GHz systems and even systems operating up to 6 GHz O2I(Outdoor-to- Indoor) losses have to be taken into accountActual measurements (eg at 38 GHz) demonstrated apenetration loss of 40 dB for brick pillars 37 dB for a glassdoor and 25 dB for a tinted glass window (indoor clear glassand drywall only had 36 and 68 dB of loss) [76] This is whyDASs are expected to be important for 5G in general and 5GIoT in particular
3GPP and ETSI propose that the pathloss incorporatingO2I building penetration loss be modelled as in the following[81]
PLb is the basic outdoor path loss where 1198893D isreplaced by 1198893D-out + 1198893D-inPLtw is the building penetration loss through theexternal wallPLin is the inside loss dependent on the depth into thebuilding and120590119875 is the standard deviation for the penetration loss
PLtw is characterized as
PL119905119908 = PL119899119901119894 minus 10 log10119873
sum119894=1
(119901119894 times 10119871119898119886119905119890119903119894119886119897 119894minus10) (3)
where
PL119899119901119894 is an additional loss is added to the external wallloss to account for non-perpendicular incidence119871119898119886119905119890119903119894119886119897 119894 = 119886119898119886119905119890119903119894119886119897 119894 +119887119898119886119905119890119903119894119886119897 119894 sdot 119891 is the penetrationloss of material 119894 example values below
119901119894 is proportion of 119894-th materials where sum119873119894=1 119901119894 = 1and119873 is the number of materials3GPP TR 38901 version 1400 Release 14ETSI TR 138 901 V1400 (2017-05)
In consideration of these propagation characteristicsmany municipalities in the US are concerned about thepossiblemassive proliferation of small cells needed to support5G For example a filing to the FCC was made in theUS late in 2018 by a consortium of towns known as theCommunities and Special Districts Coalition in responseto the Commissionrsquos September 5 2018 Draft DeclaratoryRuling and 3rd Report and Order where the FCC asserted the
claim that ldquosmall cellrdquo deployment is a federal undertakingfurthermore the filing states that ldquothe massive deploymentenvisioned by the Commission raises substantial questions asto whether the Commission is in a position to assert thatdeployment is safe given that its radio frequency emissionsrules were based on technologies and deployment patternsthat the Commission declares obsolete in this Orderrdquo [74 91]Furthermore it is unclear according to the filing what isthe size of the equipment needed to support a small cellsince it could vary from a ldquopizza boxrdquo system to severalracks that equate to 56 ldquopizza boxesrdquo [91] Although smallcells will indeed need to be deployed to properly support5G caution is advocated SampP Global Market Intelligenceestimates that small-cell deployments reach approximately850000 in the US by 2025 (with approximately 700000already deployed in 2019) with about 30 of small cellinstallations being outdoors the same projection forecasts atotal of 84 million small cells world-wide with some regionsof the world experiencing much higher deployments ratesthat in the US eg doubling the 2019 numbers by the year2025 These data show that placement within buildings is acommon alternative (there will be more in-building systemsthan outdoor systems) [75]
4 5G DAS for Indoor IoT Applications
The previous section discussed propagation issues at thehigher frequencies However even the sub-6 GHz bands haveissues penetrating buildings with the new building materialsand infrared reflecting (IRR) glass Indoor solutions areneeded for IoT even at standard 3G4G LTE frequenciesand much more so at mmWave if cellular-based (5G) IoTtransmission services for in-building applications are con-templated outdoor 5G IoT applications do not
Although it is in principle possible to support multipleaccess technologies in an IoT sensor (chipset) end-point IoTdevices tend to have low complexity in order to achieve anestablished target price point and on-board power (battery)budget Therefore a (large) number of applications will havedevices that have a single implemented wireless uplink Itfollows that -- either because of the goal of mobility support(for example a wearable that works seamlessly indoors andin open spaces around town) or because of the designerrsquos goalto utilize a single consistent IoT nodal and access technologyndash an all-sites wireless service for a Smart City application ispreferredDASsmay support such a goal (while city-wideWi-Fi andor SigfoxLoRa could be an alternative the ubiquitystandardization and cost-effectiveness of 5G cellular and IoTservices may well favor the latter in the future)
41 DAS Networks A DAS is network of a (large) numberof (small) (indoor or on-location) antennas connected to acommon cellular source via fiber optic channel providingcellularwireless service within a given structure DAS (some-times also called in-building cellular) refers to the technologythat enables the distribution and rebroadcasting of cellularLTE AWS 5G and other RF frequencies within a building orconfineddefined structural environment While DAS is oftenused in large urban office buildings DAS can also be used in
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
22 Wireless Communications and Mobile Computing
open spaces such as campuses conference centers stadiumshospitals airports train stations tunnels hotels cruise shipsand so on DASs can and will support cellular-based IoT (egLTE-MNB-IoT and 5G IoT) Elements of a DAS include (seeFigure 16)
(i) (Small) Broadband antennas and amplifiers in theindoor space (typically one or more per floor) thatshape the coverage These antennas typically coverthe entire spectrum of the cellular service (forfrommultiple service providers
(ii) Coax or fiberoptic cabling to connect the structureantennas to a local Base Station
(iii) Remote Radio Head a local Base Station (ldquosmallcellrdquo) typically in the basement and
(iv) Fiberoptic connection to an aggregation point (typ-ically in a carrier colocation space) (or the use of anoutdoor donor antenna to a specific cellular provider)The former supports carrier-neutral applications thelatter typically supports only one carrier Physicalconnectivity from the colocation space to each of thewireless providers is needed typically in the form offiber connectivity or other telecom service Businessrelationships with the wireless providers are needed
Current typical drivers include the fact that during antici-pated peak times (whether in a building or in some publicvenue as a stadium) users will experience coverage defi-ciencies blocked connections reduced data speeds amongother service deficiencies Current systems support CDMAEVDO GSM HSPA UMTS among others Future systemswill support 5G and become even more prevalent
Given themmWave transmission issues mentioned above(the small cells the directionality the free space loss andother attenuation factors) DASs will likely play a big rolein 5G both for regular voice and data services and for IoTThe large number of ldquosmall cellsrdquo cited earlier (84 million in2025 with about 70of these being considered to be indoors)supports the thesis that DASs will play a pivotal role in thefuture They will be a key element of Smart City IoT supportespecially for in-building sensors As was shown in Figure 2while a number of applications could use a Wi-Fi (or related)access technology with networked connection to the cloudor a SigfoxLoRa related solutions (these however beingvendor proprietary) Smart City IoT service implementersmay prefer to use a cellular service such as LTE-M or NB-IoTin the immediate future and 5G IoT as it becomes availableallowing a seamless and single-technology solution city-wideIn some cases for example in smallerolder buildings andorin suburbia andor for buildings very close to a 5G cell towera direct 5G IoT connection may suffice But for high-densityurban and smart building applications the use of DASs seemsinevitable
42 DAS Design A single carrier-neutral consolidated sys-tem is often sought a carrier-neutral system avoids mul-tiplicity of antenna distribution and sharing allows morecoverage and higher capacity A carrier-neutral DAS supportsan end-use system for example a smartphone regardless of
which service provider the user is subscribed to It would berather expensive for a building owner to deploy a carrier-neutral DAS that supports a single building unless it wouldbe a very large building campus or installation Withcarrier-neutral DAS arrangements the ownership of systemis shifted from the building owner or a specific cellularcarrier to a third-party system provider or a DAS integra-tor Figure 17 depicts a typical carrier-neutral arrangementObtaining wireless carrier permission and coordinatingbetween different wireless carriers is a key planning under-taking of any successful DAS rollout Three scenarios areshown
(i) ScenarioApproach S1 The DAS integratorproviderwires up a remote building or space and drops a fiberlink into an existing colo rack at an existing carrier-neutral provider thus sharing all the Base StationHotel (BSH) colo equipment and interfaces to thevarious wireless providers
(ii) ScenarioApproach S2 The DAS integratorprovidermust build out the requisite base station equipmentin the colo (the colo provider only provides powerrack space HVAC and so on) The DAS integra-torprovider must also build interfaces to the wirelessproviders and secure business arrangements withthem The DAS integratorprovider builds out theremote buildings or venues
(iii) ScenarioApproach S3 The DAS integratorprovidermust build out the requisite base station equipmentin the colo but the DAS integratorprovider canmakeuse of existing interfaces and equipment to the vari-ous wireless providers The DAS integratorproviderbuilds out the remote buildings or venues
A less desirable approach is to use ldquodonor antennasrdquo (alsoshown in Figures 16 and 17) These antennas are installedon the roof of a building and are pointed at ldquodonorrdquo celltowers Typically a single cellular vendor is supported Thein-building arrangement is similar to that of a carrier-neutralarrangement except that there typically will not be a remotebase station a combination of fiber optic cable coaxial cableand in-building antennas is used to amplify and distributethose signals within a given space coordination with thegiven carrier is still needed tomake sure that the concentratedtraffic is accepted by the provider
5 5G Deployment Snapshot
51 5G Cellular Services According to GSMA 5G is on trackto account for 15 (14 billion) of global mobile connectionsby 2025 By early 2019 according to GSMA eleven worldwideoperators had announced initial 5G service launches andseven other operators had activated 5G base stations withcommercial services to follow in the near future [92] Selectcities worldwide will have 5G by the end of 2019 See Table 6for a summary of near-term 5G service-deployment activi-ties However 4G services are expected to continue well past2025 4G will account for 59 of the connections 3G for 20of the connections and 2G 5 of the connection (3G and 2G
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 23
Figure 16 Elements of a DAS
Figure 17 Carrier-neutral DAS
are trending down through this periodwhile 4Gwill continueto grow but settle at around 60 by 2023 -- 5G is penetratingat approximate linear rate of CAGR around 15-2) Alsoaccording to GSMA 2019 will see 5G launches accelerate anddevices hit the market16 major markets worldwide will startto offer commercial 5G networks in 2019 following on fromthe first 5G launches in South Korea and the US in 2018 asfollows [28]
(i) Q4 2018 South Korea US
(ii) Q1 2019 Bahrain Czech Republic Estonia FinlandSaudi Arabia Switzerland
(iii) Q2 2019 Australia Qatar(iv) Q3 2019 Austria China Hong Kong Kuwait Spain
UAE(v) Q4 2019 Portugal UK
As of Q2 2019 there were 303 rollouts of 5Gmobile networksacross 294 locations worldwide operated by 20 mobile
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
24 Wireless Communications and Mobile Computing
Table 6 Summary of near-term 5G service-deployment activities (2019 view)
Country or Region Near-term 5G Activities
South Korea
Korea Telecom rolled out a trial 5G network in support of the 2018 Winter Olympics in SeoulSouth Korea covering events in several cities It has also launched a VRAR games platformsupported from the cloud over 5G SK Telecom has acquired spectrum in the 35 GHz and 28 GHzfrequencies in preparation of deploying 5G
China
China plans early implementations of 5G The GSMA estimates that by 2025 China will represent40 percent of global 5G connections According to the GSMA with 460 million projected usersChina is expected to become the worldrsquos largest 5G market by 2025 higher than Europersquos 205million and the United Statesrsquo 187 million combined Chinarsquos three major mobile operators -China Mobile China Unicom and China Telecom - are rolling out trial operations of 5G systemsin several cities and all three aim to fully commercialize the technology by 2020 [66]
India 5G trials are contemplated by late 2019 and early deployments may happen late in 2020
Japan NTT DoCoMo demonstrated an advanced security service based on 5G network technology foruse in the 2020 Olympics
United StatesMigration from the 4G networks should be relatively simple The FCC has been making severalnew bands available as noted elsewhere Verizon has been aggressive in its advertisementcampaigns about its introduction of 5G-related services
EuropeT-Mobile is preparing for the rollout of 5G in 2020 starting in the Netherlands Some fear thatEurope risks falling behind other regions because of restrictive regulation and weak investmentsless than half of the countries in Europe have actually allocated spectrum for 5G [67]
carriers [93] In the US 21 deployments were documentedof which five were in Texas four in California two eachin North Carolina and Florida and one each in OklahomaMinnesota Illinois Indiana Kentucky Tennessee Georgiaand Louisiana (some of these such as the ATampT 5G networkin Louisville KY had ldquoLimited Availabilityrdquo at that time)
52 MmWave Spectrum Regarding frequency bands 3GPPis initially focusing on 24 GHz to 43 GHz mmWave spec-trum (Release 15) Other ongoing 5G work relates to NSAand SA configurations Massive MIMO beamforming andLTE interoperability 3GPP Release 16 (2019) aims at fullcompliance to IMT-2020 (eg supporting 1 GHz channels)and other spectrum capabilities (eg spectrum sharingadditional bands and URLCC)
In the US among other possible candidates the FCCis making available new frequency bands for 5G use underits rubric of ldquoSpectrum Frontiers proceedingrdquo of which threehave been instituted in the recent past With the ldquoJuly 2016Orderrdquo the FCC designated the 275-2835 GHz (knownas the ldquo28 GHz bandrdquo) 37-386 GHz (known as the ldquo37GHz bandrdquo) and 386-40 GHz (known as the ldquo39 GHzbandrdquo) bands for flexible mobile and fixed commercial useand designated the 64-71 GHz band for unlicensed use (tosupplement 57-64 GHz which had been made available forunlicensed use at an earlier time) While the FCC has yet toauction any of the newUpperMicrowave Flexible Use Serviceor (UMFUS) spectrum in 2017 with a Second Report andOrder a Memorandum Opinion and Order it designated anadditional 1700megahertz of mmWave spectrum for licensedflexible commercial wireless fixed and mobile use The 1700MHz spectrum covered the 2425-2445 2475-2525 and472-482 GHz bands (the first two known collectively as theldquo24 GHzrdquo bands and the third known as the ldquo47 GHzrdquoband) Therefore the spectrum at 2425-2445 GHz is nowallocated for non-Federal fixed and mobile services on a
co-primary basis and the spectrum at 2475-2525 GHz fornon-Federal fixed mobile and fixed-satellite (FSS) serviceson a co-primary basis [94]
In terms of rollouts in the US the spectra at 275 ndash2835 GHz and 37 ndash 40 GHzmay see preliminary commercialdeployments in 2019 in Korea the spectrum at 265 ndash 295GHz is similarly expected to see commercial deployments in2019 and the EU expects commercial deployments for the2425 ndash 275 GHz spectrum starting around 2020
In addition to the radio access for the end-user devicethere is also interest in Backhaul and now also in FronthaulBackhaul mechanisms are mechanisms to connects the wire-less network to the wired network by backhauling traffic fromdispersed cell sites toMobile SwitchingOffices (MSOs)Theselinks typically are either traditional transmission systems(such as SONET or point-to-point microwave at variousoperating bands) or they are Ethernet-over-Fiber links (eg1 GbE or 10 GbE) A UMa site has Baseband Unit (BBU) thatprocesses user and control data which is in turn connected toa Radio Unit (RU) to generate radio signals transmitted overthe air via the tower-mounted antennas
Fronthaul is related to a new type of Radio AccessNetwork (RAN) architecture that is comprised of centralizedbaseband controllers and standalone radio heads installedat remote UMa or UMi sites possibly many miles away Inthe fronthaul model the BBU and RU equipment is locatedfurther away from each other than is the case in the backhaulmodel The RU equipment (now referred to as a RemoteRadio Head [RRH]) is still located at the cell site but the BBUis relocated to centralized location where it supportsmultipleRRHs See Figure 18 The optical links that interconnectthe newly centralized BBU and the multiple RRHs is arereferred to as fronthaul The use of fronthaul-based C-RAN(Cloud-RAN) architectures typically improves the cell edgeperformance Backhaul and fronthaul are key use cases formmWave spectrum and will play a role in 5G and 5G IoT
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 25
Remote RadioHead (RRH)
Remote RadioHead (RRH)
CPRILinks
CPRILinks
CPRI Interface
BasebandUnit (BBU)
MCO
DWDM OTN Network
Fronthaul Backhaul
DWDM OTN link
Figure 18 Fronthaul and Backhaul
A consortium of wireless equipment vendors standard-ized the Common Public Radio Interface (CPRI) protocolthat runs over these fronthaul links a few years ago morerecently a newer eCPRI 10 interface has been defined addi-tionally work is underway to defining a more detailed inter-face The tight performance requirements of CPRIeCPRI-- capacity distance and latency ndash drive towards fiberconnectivity such as DWDM (or more specifically OTN[Optical Transport Network]) systems between centralizedBBUs and the RRHs Ethernet-based solutions have existedfor a number of years using mmWave spectrum Work isunderway in 3GPP to define backhauling solutions using thesame spectrum as access Work is also underway to definenew fronthaul interfaces also utilizing mmWave spectrum
53 5G IoT Services Global IoT revenue are expected toincrease at an annual rate of 23 to 2025 to reach $11 trillion(up from 267 B in 2018) As discussed in the next sectionnear term ldquo5G IoTrdquo really equate to NB-IoT and LTE-Mcapabilities At the end of 2018 there were 83 commercialdeployments of LTE-M and NB-IoT worldwide Howeverpure connectivity will become increasingly commoditizedmaking it difficult for operators to compete on the datatransmission alone declining from 9 of total IoT revenuein 2018 to 5 in 2025 Service providers must developnew strategies and business models beyond connectivityservices Applications platforms and services (eg cloud dataanalytics and IoT security) are the major growth areas of IoTthis segmentwill be approximately 70 of themarket in 2025Professional services (eg consulting systems integrationalso including managed services) will increase in share andwill be approximately 25 of the market in 2025 [28]
6 Current Alternatives and Convergence to 5G
5G IoT will need to compete with other technologies bothof the cellular type (eg NB-IoT and LTE-M) as well asthe non-cellular type (although NB-IoT and LTE-M are nowconsidered ldquopart of the 5G worldrdquo) The economics and
availability of these ldquolegacyrdquo networks in various parts ofthe world may be such that a level of inertia frustrating afull migration to truly-novel 5G IoT services will take holdClearly in principle 5G is better positioned for cityregion-wide applications as contrasted with building or campusapplications
From an end-user perspective design and implementa-tion questions center around the following issues which 5GIoT technology must be able to address successfully
(i) Availability of equipment(ii) Availability of service (geographic coverage in the
area of interest)(iii) Support of required technical details (latency band-
width packet loss and so on)(iv) Support of mobility (where needed eg wearables
crowdsensing Vehicle to Vehicle and Vehicle toInfrastructure applications to name a few)
(v) Adequate reliability (where needed eg physicalsecurity process control Vehicle to Vehicle and Vehi-cle to Infrastructure applications to name a few)
(vi) Scalability support (functional and geographicnu-merical expansion of the application)
(vii) Initial and recurring cost of the equipment and(viii) Initial and recurring cost of the service
Recent acceptability and economics of NB-IoT and LTE-Mcan serve as a proxy for the near-term commercial successof 5G IoT in particular and truly-novel 5G IoT services ingeneral Some developers have looked at cellular services forcity-wide or region-wide IoT coverage in some instances forexample for national truck transportation a combination ofLow Earth Orbit (LEO) satellite service and cellular serviceshave and are being used A current drawback is the costof the requisite (miniaturized) modems and the cost of thecellular service New services such as NB-IoT and LTE Cat-M1 (an LTE-based 3GPP-sponsored alternative to NB-IoT
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
26 Wireless Communications and Mobile Computing
also known as LTE-M) are short term attempts to addressthe cost and resource issues In particular NB-IoT is seen asproviding a pathway to 5G IoT 5G and truly-novel 5G IoT arethe target solutions
61 NB-IoT As noted earlier NB-IoT is a licensed low powerLPWAN technology designed to coexist with existing LTEspecifications and providing cellular-level QoS connectivityfor IoT devices NB-IoT was standardized by 3GPP in LTERelease 13 but it does not operate in the LTE context perse [95ndash97] NB-IoT has attracted support from QualcommEricsson andHuawei amongmany other vendors and serviceproviders NB-IoT (also known as LTE Cat-NB1) is based ona Direct Sequence Spread Spectrum (DSSS) modulation ina 200 kHz channel There are several underutilized 200-kHzGSM spectrum channels as well as other possible bands suchas guard bands NB-IoT is intended as an alternative to LoRaand Sigfox This technology can optimize sunken financialinvestments by service providers and can shorten the servicedeployment rollout timetable for IoT services since NB-IoTuses existing cellular infrastructure NB-IoT service goalsinclude (i) low complexity end-nodes (ii) device cost lessthan $5 (iii) a device battery life expected to last for 10 years ifit transmits 200 bytes of data per day and (iv) uplink latencyless than 10s (thus not a true real-time service) NB-IoToperates on 900-1800 MHz frequency bands with coverageof up to about 20 miles it supports data rates of up to 250Kbps for uplink and 230 Kbps for downlink communications[98ndash101] NB-IoT can be implemented in a number of ways(i) in standalone non-cellular licensed bands (ii) in unused200 kHz bands in the context of GSM or CDMA and (iii) inLTE environments where base stations can allocate a resourceblock to NB-IoT transmissions Since NB-IoT offers low costfor the device and for the service it is a good choice for large-scale distributed deployment in Smart Cities and smart gridapplications
As illustrative commercial examples in 2018 T-Mobileannounced a North American NB-IoT plan that costs just $6a year ndash one tenth of Verizonrsquos Cat-M plansndash for up to 12 MBper connected device and several NB-IoT modules based onQualcomm MDM9206 LTE IoT modem that are certifiedfor use on T-Mobilersquos network T-Mobile in conjunction withQualcomm and Ericsson conducted the first trial NB-IoT inthe US in 2017 across multiple sites T-Mobile and the Cityof Las Vegas also announced a partnership to deploy IoTtechnology throughout the city For applications that requiremore bandwidth and voice T-Mobile offers Cat-1 IoT AccessPacks [102 103] NB-IoT consumes minimal power whilemost IoT end-nodes save power when they are quiescentwhen the node and the modem are running and handlingall the signal processing the systems with simpler waveform(such NB-IoT) consume less overall power Additionallychipsets that support a single protocol (such as NB-IoT)are cheaper compared to a chipset that supports multipleprotocols Furthermore prima facie NB-IoT may providedeeper building penetration than LTE-M
62 LTE-M LTE-M is a power-efficient system where twoinnovations support battery efficiency LTE eDRX (Extended
Discontinuous Reception) and LTE PSM (Power SavingMode) LTE-M allows the upload of 10 bytes of data aday (LTE-M messages are fairly short compared to NB-IoT messages) but also allows access to Mbps rates There-fore LTE-M can support several use cases In the USmajor carriers such as Verizon and ATampT offer LTE-Mservices (as noted Verizon has announced support for NB-IoT -- T-Mobile and Sprint appears to lean in the NB-IoT direction) [104] Worldwide geographies with GSMdeployments will likely offer NB-IoT in the short termFigure 19 depicts some of the IoT compatibility mechanismsto be incorporated into 5G in terms of band and band-width however the transmission frequencies will be wildlydifferent
In summary LTE-M supports low nodal complexityhigh nodal density low nodal power consumption lowlatency and extended geographic coverage while allowingservice operators the reuse of the LTE installed base NB-IoT aims at improved indoor coverage high nodal densityfor low throughput devices low delay sensitivity low nodecost low nodal power consumption and simplified networkarchitecture NB-IoT and LTE-M are currently providingmobile IoT solutions for smart cities smart logistics andsmart metering but only in small deployments to date (asof early 2018 there were 43 commercial NB-IoT and LTE-Mnetworks worldwide [105]) As noted the commercial successof NB-IoT and LTE-M can serve as a proxy for the eventualsuccess of 5G IoT in a smart city context (comparedwith non-cellular LPWAN solutions)
NB-IoT LTE-M and LTE are 4G standards but advocatesclaim that they remain integral parts of early releases of5G Proponents make the case that ldquoenterprises deployingeither NB-IoT or LTE-M are futureproofing their IoT projectsbecause when 5G rollouts become commonplace these twoMobile IoT standards will continue into foreseeable 5G releases(from 3GPP Release 15 on)rdquo [102] In the context of 3GPPRel 15 it appears in fact that NB-IoT and LTE-M will beincluded as 5Gmobile standards In 2018 the GSMA assertedthat ldquoNB-IoT and LTE-M as deployed today are part ofthe 5G family with the dawn of the 5G era [] both NB-IoT and LTE-M technologies are an integral part of 5G andthat 5G from the LPWA perspective is already here todayrdquo[105] Including these technologies as initial 5G IoT standardswill motivate service providers and vendors to support theseimplementations for IoT deployments as an evolutionarystrategy to 5G 3GPP Release 16 (targeted for the end of2019) is considered to be the ldquosecond 5G standardrdquo andthereafter transmitted to the ITU for consideration as a globalstandard Among other functionality and capabilities Release16 is expected to add standards for connected cars and smartfactories (notably automobile companies have formed the 5GAutomotive Association to assist 3GPP to set autonomousvehicle standards such as 5G cellular vehicle-to-everything[C-V2X])
7 Conclusion
This paper discussed a number of issues related to 5G-basedIoT applications particularly in Smart Cities environments
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 27
Frequency
Current (sub 1 GHz) 5G (above 6 GHz and likely in the 24+ GHz)
LTE
LTE-
M
NB-
IoT
5G NR
LTE-
M
NB-
IoT
Figure 19 Support of LTE-M and NB-IoT under 5G
including the need for small cells the transmission issues atthe millimeter wave frequencies building penetration issuesthe need for DAS and the near term introduction of pre-5GIoT technologies such as NB-IoT and LTE-M these beingpossible proxies for 5G IoT deployment
A firm definition of 5G IoT has still to emerge althougha large number of use cases have been described by variousindustry entities Both 3GPP NB-IoT and LTE-M technolo-gies are seen at this juncture as integral to 5G servicesthese 4G technologies are expected to continue under fullsupport in 5G networks for the immediate future HoweverIoTSmart City applications that require high bandwidth willneed implementations of eMBB and mmWave frequencies
Some controversy existed at press time about the devel-opment of 5G equipment in the context of origin-of-manufacturing and the possible intrinsic risk related tocybersecurity [106] If these issues are not satisfactorilyresolved somedelay in the broad early deployment of 5Gmayresult However the expectation is that these issues will workthemselves out over time
Conflicts of Interest
The authors declare that they have no conflicts of interest
References
[1] National League of Cities and Center for City Solutionsand Applied Research ldquoCity of the future ndash technology andmobilityrdquo White Paper 1301 Pennsylvania Avenue Suite 550Washington DC USA 2015
[2] A Ramaswami A G Russell P J Culligan K Rahul Sharmaand E Kumar ldquoMeta-principles for developing smart sustain-able and healthy citiesrdquo Science vol 352 no 6288 pp 940ndash9432016
[3] D R Martınez T J Gracia E M Munoz and A C GarcıaldquoSmart citiesrsquo challenge how to improve coordination in thesupply chainrdquo in Sustainable Smart Cities Innovation Tech-nology and Knowledge Management pp 129ndash142 SpringerInternational Publishing 2016
[4] N Mali ldquoA review on smart city through internet of things(IoT)rdquo International Journal of Advanced Research in ScienceManagement and Technology vol 2 no 6 2016
[5] A Caragliu C D Bo and P Nijkamp ldquoSmart cities in europerdquoJournal of Urban Technology vol 18 no 2 pp 65ndash82 2011(Chinese)
[6] D Minoli and B Occhiogrosso ldquoInternet of things applicationsfor smart citiesrdquo in Internet of Things A to Z Technologies and
Applications Q Hassan Ed Chapter 12 IEEE PressWiley2018
[7] A Zanella N Bui A P Castellani L Vangelista and M ZorzildquoInternet of things for smart citiesrdquo IEEE Internet of ThingsJournal vol 1 no 1 pp 22ndash32 2014
[8] D Minoli and B Occhiogrosso ldquoMobile IPv6 protocols andhigh efficiency video coding for smart city IoT applicationsrdquoin Proceedings of the 13th International Conference and Expo onEmerging Technologies for a Smarter World (CEWIT) pp 1ndash6Stony Brook New York NY USA 2017
[9] B J Wewalaarachchi H Shivanan and H GunasinghamldquoIntegration platform to enable operational intelligence anduser journeys for smart cities and the internet of thingsrdquo inProceedings of the Patent US20160239767 A1 2016
[10] S Srivastava and N Pal ldquoSmart cities the support for internetof things (IoT)rdquo International Journal of Computer Applicationsin Engineering Sciences pp 5ndash7 2016
[11] O Bates and A Friday ldquoBeyond data in the smart cityrepurposing existing campus IoTrdquo IEEE Pervasive Computingvol 16 no 2 pp 54ndash60 2017
[12] D Kyriazis T Varvarigou D White et al ldquoSustainable smartcity IoT applications heat and electricity management amp eco-conscious cruise control for public transportationrdquo in Proceed-ings of the IEEE 14th International Symposium on ldquoA World ofWireless Mobile andMultimedia Networksrdquo (WoWMoM) IEEEMadrid Spain 2013
[13] D Minoli and B Occhiogrosso ldquoIoT applications to smartcampuses and a case studyrdquo EuropeanUnionDigital Library vol5 article e4 pp 2518ndash3893 2017
[14] A Al-Fuqaha M Guizani M Mohammadi et al ldquoInternetof things a survey on enabling technologies protocols andapplicationsrdquo IEEE Communication Surveys ampTutorials vol 17no 4 pp 2347ndash2376 2015
[15] R Gomes H Pombeiro C Silva et al ldquoTowards a smartcampus building-user learning interaction for energy effi-ciency the lisbon case studyrdquo in Handbook of Theory andPractice of Sustainable Development in Higher Education WorldSustainability Series pp 381ndash398 Springer 2016
[16] Z Yu Y Liang B Xu et al ldquoTowards a smart campus withmobile social networkingrdquo in Proceedings of the 4th IEEE IntrsquolConference on Cyber Physical and Social Computing (CPSCom)pp 162ndash169 IEEE Dalian China 2011
[17] A Roy J Siddiquee A Datta et al ldquoSmart traffic amp parkingmanagement using IoTrdquo in Proceedings of the IEEE 7th AnnualInformation Technology Electronics andMobile CommunicationConference (IEMCON) IEEE Vancouver BC Canada 2016
[18] R Grodi D B Rawat and F Rios-Gutierrez ldquoSmart parkingParking occupancy monitoring and visualization system for
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
28 Wireless Communications and Mobile Computing
smart citiesrdquo in Proceedings of the SoutheastCon IEEE NorfolkVA USA 2016
[19] D Minoli K Sohraby and B Occhiogrosso ldquoIoT consider-ations requirements and architectures for smart buildings ndashenergy optimization and next generation buildingmanagementsystemsrdquo IEEE Internet of Things Journal vol 4 no 1 pp 269ndash283 2017
[20] L Kang S Poslad W Wang et al ldquoA public transport bus asa flexible mobile smart environment sensing platform for IoTrdquoin Proceedings of the 12th International Conference on IntelligentEnvironments (IE) IEEE London UK 2016
[21] M Alam J Ferreira and J Fonseca ldquoIntroduction to intelligenttransportation systemsrdquo in Journal of Intelligent TransportationSystems vol 52 of Studies in Systems Decision and Control pp1ndash17 Springer International Publishing 2016
[22] J Barbaresso G Cordahi and et al ldquoThe national academiesof science engineering and medicine USDOTrsquos intelligenttransportation systems (ITS) its strategic planrdquo The NationalAcademies of Science Engineering and Medicine USDOTrsquosIntelligent Transportation Systems (ITS) ITS Strategic Plan2015-2019 FHWA-JPO-14-145 2014
[23] S H Sutar R Koul and R Suryavanshi ldquoIntegration of SmartPhone and IOT for development of smart public transportationsystemrdquo in Proceedings of the International Conference onInternet of Things and Applications (IOTA) pp 73ndash78 PuneIndia 2016
[24] M Picone S Busanelli M Amoretti et al Advanced Technolo-gies for Intelligent Transportation Systems Springer 2015
[25] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the Global Internet of ThingsSummit (GIoTS) IEEE Geneva Switzerland 2017
[26] D Minoli and B Occhiogrosso ldquoUltrawideband (UWB) tech-nology for smart cities IoT applicationsrdquo in Proceedings of theIEEE International Smart Cities Conference (ISC2) pp 1ndash8IEEE Kansas City Mo USA 2018
[27] R K Ganti F Ye and H Lei ldquoMobile crowdsensing currentstate and future challengesrdquo IEEE Communications Magazinevol 49 no 11 pp 32ndash39 2011
[28] Global System for Mobile Communications Association(GSMA)TheMobile Economy 2019 2019 httpswwwgsmain-telligencecomresearchfile=b9a6e6202ee1d5f787cfebb95d36-39c5ampampdownload
[29] Statistica Internet of things (IoT) connected devices installed baseworldwide from 2015 to 2025 (in billions) 2015 httpswwwstatistacomstatistics471264iot-number-of-connected-devi-ces-worldwide
[30] ldquoRecommendation ITU-R M2083-0 IMT visionmdashframeworkand overall objectives of the future development of IMT for2020 and beyondrdquo httpwwwituintrecR-REC-M2083-0-201509-I
[31] ldquoNext generation mobile networksrdquo 5G White Paper 2015httpswwwngmnorgfileadminngmncontentdownloadsTechnical2015NGMN 5G White Paper V1 0pdf
[32] ldquoFeasibility study on new services and markets technologyenablersrdquo 3GPP 22891 2019 httpportal3gpporgdesk-topmodulesSpecificationsSpecificationDetailsaspxspecifica-tionId=2897
[33] H Yu H Lee and H Jeon ldquoWhat is 5G emerging 5G mobileservices andnetwork requirementsrdquo Sustainability vol 9 no 10article 1848 2017
[34] GSMA Road to 5G introduction and migration 2018 httpswwwgsmacomfuturenetworkswp-contentuploads201804Road-to-5G-Introduction-and-Migration FINALpdf
[35] S W Hu and C M Shy ldquoHealth effects of waste incinerationa review of epidemiologic studiesrdquo Journal of the Air amp WasteManagement Association vol 51 no 7 pp 1100ndash1109 2001
[36] A Santarsiero G Trevisan G Cappiello et al ldquoUrban cremato-ria emissions as they stand with current practicerdquoMicrochemi-cal Journal vol 79 no 1-2 pp 299ndash306 2005
[37] M Takaoka K Oshita N Takeda and S Morisawa ldquoMercuryemission from crematories in Japanrdquo Atmospheric Chemistryand Physics vol 10 no 8 pp 3665ndash3671 2010
[38] N Takeda M Takaoka K Oshita and S Eguchi ldquoPCDDDFand co-planar PCB emissions from crematories in JapanrdquoChemosphere vol 98 pp 91ndash98 2014
[39] Y Xue H Tian J Yan et al ldquoPresent and future emissions ofHAPs from crematories in Chinardquo Atmospheric Environmentvol 124 pp 28ndash36 2016
[40] D Muenhor J Satayavivad W Limpaseni et al ldquoMercurycontamination and potential impacts from municipal wasteincinerator on Samui Island Thailandrdquo Journal of Environmen-tal Science and Health Part A ToxicHazardous Substances andEnvironmental Engineering vol 44 no 4 pp 376ndash387 2009
[41] S Sakai K Hayakawa H Takatsuki and I Kawakami ldquoDioxin-like PCBs released fromwaste incineration and their depositionfluxrdquo Environmental Science amp Technology vol 35 no 18 pp3601ndash3607 2001
[42] G D Hinshaw and A R Trenholm ldquoHazardous waste inciner-ation emissions in perspectiverdquoWaste Management vol 21 no5 pp 471ndash475 2001
[43] D C Ashworth G W Fuller M B Toledano et al ldquoCom-parative assessment of particulate air pollution exposure frommunicipal solid waste incinerator emissionsrdquo InternationalJournal of Environmental Research and Public Health vol 201313 pages 2013
[44] Chapter 4 in Waste Incineration amp Public Health NationalResearch Council (US) Committee on Health Effects of WasteIncineration Washington (DC) National Academies Press (US)2000 ISBN-10 0-309-06371-X Also at httpswwwncbinlmnihgovbooksNBK233615
[45] S Bose-OrsquoReilly K M McCarty N Steckling et al ldquoMercuryexposure and childrenrsquos healthrdquo Current Problems in Pediatricand Adolescent Health Care vol 40 no 8 pp 186ndash215 2010
[46] G Gonzalez-Cardoso N Santiago J M Hernandez-Contrerasand M Gutierrez ldquoPM25 emissions from urban crematori-umsrdquo Energy Procedia vol 153 pp 359ndash363 2018
[47] METIS mobile and wireless communications enablers forthe twenty-twenty (2020) Information society the 5G futurescenarios identified by METIS ndashthe first step toward A 5Gmobile and wireless communications system 2013
[48] A Osseiran V Braun T Hidekazu et al ldquoThe foundationof the mobile and wireless communications system for 2020and beyond challenges enablers and technology solutionsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring) IEEE Dresden Germany 2013
[49] ICT-317669 METIS project ldquoRequirements and general designprinciples for new air interfacerdquo httpswwwmetis2020comdocumentsdeliverables 2013
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Wireless Communications and Mobile Computing 29
[51] ICT-317669 METIS project ldquoIntermediate description of thespectrum needs and usage principlesrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[52] ICT-31766 METIS project ldquoSummary on preliminary trade-offinvestigations and first set of potential network-level solutionsrdquohttpswwwmetis2020comdocumentsdeliverables 2013
[53] ICT-317669 METIS project ldquoNovel radio link conceptsand state of the art analysisrdquo httpswwwmetis2020comdocumentsdeliverables 2013
[54] ICT-317669METIS project ldquoComponents of a new air interface- building blocks and performancerdquo httpswwwmetis2020comdocumentsdeliverables 2014
[55] ldquoSmall cell forum mmWave 5G eMBB use cases amp small cellbased hyperdense networksrdquo 2017
[56] X Ge L Pan Q Li et al ldquoMultipath cooperative communica-tions networks for augmented and virtual reality transmissionrdquoIEEE Transactions onMultimedia vol 19 no 10 pp 2345ndash23582017
[57] G Tech Y Chen K Muller et al ldquoOverview of the multiviewand 3D extensions of high efficiency video codingrdquo IEEETransactions on Circuits and Systems for Video Technology vol26 no 1 pp 35ndash49 2016
[58] J Horwitz ldquoFCC expands 35GHz band to 5G and opens 6 GHzband to future Wi-Firdquo httpsventurebeatcom20181023fcc-expands-3-5ghz-band-to-5g-and-opens-6ghz-band-to-fu-ture-wi-fi 2018
[59] D Minoli K Sohraby and B Occhiogrosso ldquoIoT security(IoTSec) mechanisms for e-health and ambient assisted livingapplicationsrdquo in Proceedings of the 2nd IEEE InternationalConference on Connected Health Applications Systems andEngineering Technologies (CHASE) IEEE Philadelphia PaUSA 2017
[60] D Minoli K Sohraby and J Kouns ldquoIoT Security (IoTSec)Considerations Requirementsrdquo in Proceedings of the 14th IEEEAnnual Consumer Communications amp Networking Conference(CCNC) IEEE Las Vegas NV USA 2017
[61] ldquoSecurity architecture and procedures for 5G Systemrdquo 3GPP TS33501 2018 httpwww3gpporgDynaReport33501htm
[62] ldquoStudy on the security aspects of the next generation sys-temrdquo 3GPPTR 33899 2017 httpwww3gpporgDynaReport33899htm
[63] ldquo5G Enablers for Network and System Security and Resiliencerdquohttpwww5gensureeu 2017
[64] GDPR General Data Protection Regulation European Union2016 httpeur-lexeuropaeulegal-contentenALLuri=CELEX32016R0679
[65] ePrivacy Directive on privacy and electronic communica-tions European Union 2002 httpseur-lexeuropaeulegal-contentenALLuri=CELEX32002L0058
[66] The Strait Times Staff China to Be Biggest 5G Marketby 2025 Report 2019 httpswwwstraitstimescomworldchina-to-be-biggest-5g-market-by-2025-report
[67] J Stubbs andD BusvineWeak investment climate main 5G risknot security fears Ericsson httpswwwreuterscomarticleus-telecoms-mobileworld-ericssonweak-investment-climate-main-5g-risk-not-security-fears-ericsson-idUSKCN1QE0ST
[68] FCC Millimeter Wave Propagation Spectrum ManagementImplications vol 70 Federal Communications CommissionOffice of Engineering and Technology New Technology Devel-opment Division Washington DC USA 1997
[69] P Tracy What is Mm Wave and How Does It Fit into 5G2016 httpswwwrcrwirelesscom20160815fundamentalsmmwave-5g-tag31-tag99
[70] X ZhangMillimeterWave for 5G UnifyingCommunication andSensing 2015 httpswwwmicrosoftcomen-usresearchwp-contentuploads201503Xinyu-Zhang 5GmmWavepdf
[71] X Ge J Yang H Gharavi and Y Sun ldquoEnergy efficiencychallenges of 5G small cell networksrdquo IEEE CommunicationsMagazine vol 55 no 5 pp 184ndash191 2017
[72] XGe Z Li and S Li ldquo5G software defined vehicular networksrdquoIEEE Communications Magazine vol 55 no 7 pp 87ndash93 2017
[73] S Sun T S Rappaport S Rangan et al ldquoPropagation path lossmodels for 5G urban micro- and macro-cellular scenariosrdquo inProceedings of the 83rd IEEE Vehicular Technology Conference(VTC Spring) IEEE Nanjing China 2016
[74] ldquoDraft declaratory ruling and third report and orderrdquo WCDocket No 17-84 WT Docket No17-79 FCC-CIRC1809-022018
[75] ldquoSmall cell forumrdquo Small Cells Market Status Report 2018httpwwwscfioendocuments050 Small cells market sta-tus report February 2018phputm source=Emailcampaignamputm medium=eshotsamputm campaign=membereshot
[76] T S Rappaport Y Xing G R MacCartney et al ldquoOverviewof millimeter wave communications for fifth-generation (5G)wireless networks-with a focus on propagation modelsrdquo IEEETransactions on Antennas and Propagation vol 65 no 12 pp6213ndash6230 2017
[77] T Rappaport S Sun R Mayzus et al ldquoMillimeter wave mobilecommunications for 5G cellularrdquo IEEE Access vol 1 pp 335ndash349 2013
[78] T Bai A Alkhateeb and R W Heath ldquoCoverage and capacityof millimeter-wave cellular networksrdquo IEEE CommunicationsMagazine vol 52 no 9 pp 70ndash77 2014
[79] S Rangan T S Rappaport and E Erkip ldquoMillimeter-wave cel-lular wireless networks potentials and challengesrdquo Proceedingsof the IEEE vol 102 no 3 pp 366ndash385 2014
[80] W Roh J-Y Seol J Park et al ldquoMillimeter-wave beamformingas an enabling technology for 5G cellular communications the-oretical feasibility and prototype resultsrdquo IEEECommunicationsMagazine vol 52 no 2 pp 106ndash113 2014
[81] 3GPP 38901-e20 ldquo3rd generation partnership projectrdquo Tech-nical Specification Group Radio Access Network Study onChannel Model For Frequencies From 05 to 100 GHz (Release14) 2017
[82] F Khan and Z Pi ldquommWave mobile broadband (MMB)unleashing the 3ndash300GHz spectrumrdquo in Proceedings of the 34thIEEE Sarnoff Symposium pp 1ndash6 Princeton NJ USA 2011
[83] Z Pi and F Khan ldquoAn introduction to millimeter-wave mobilebroadband systemsrdquo IEEE Communications Magazine vol 49no 6 pp 101ndash107 2011
[84] F Khan and Z Pi ldquoMillimeter-wave mobile broadbandunleashing 3-300 GHz spectrumrdquo in Proceedings of the IEEEWireless Communications and Networking Conference IEEE2011
[85] S Rajagopal S Abu-Surra Z Pi and F Khan ldquoAntenna arraydesign for multi-Gbps mmwave mobile broadband communi-cationrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM) pp 1ndash6 Houston Tex USA 2011
[86] A Ghosh T A Thomas M C Cudak et al ldquoMillimeter-wave enhanced local area systems a high-data-rate approachfor future wireless networksrdquo IEEE Journal on Selected Areas inCommunications vol 32 no 6 pp 1152ndash1163 2014
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
30 Wireless Communications and Mobile Computing
[87] G R MacCartney and T S Rappaport ldquoStudy on 3GPPrural macrocell path loss models for millimeter wave wirelesscommunicationsrdquo in Proceedings of the ICC 2017 - 2017 IEEEInternational Conference on Communications pp 1ndash7 IEEEParis France 2017
[88] Y-S Lu C-F Lai C-C Hu and Y-M Huang ldquoPath lossexponent estimation for indoor wireless sensor positioningrdquoKSII Transactions on Internet and Information Systems vol 4no 3 article 243 2010
[89] S Srinivasan and M Haenggi ldquoPath loss exponent estimationin largewireless networksrdquo InformationTheory andApplicationsWorkshop pp 124ndash129 2009
[90] M Viswanathan Log Distance Path Loss or Log Normal Shad-owing Model 2013 httpswwwgaussianwavescom201309log-distance-path-loss-or-log-normal-shadowing-model
[91] G L Lederer ldquoSmart communities and special districts coali-tion ndash ex parte submission accelerating wireless broadbanddeployment by removing barriers to infrastructure investmentWT docket no 17-79 and no 17-84rdquo Best Best amp Krieger LLP2018 2000 Pennsylvania AvenueNW Suite 5300WashingtonDC 20006
[94] C Yorkgitis FCC Adopts a Second Wave of Millimeter WaveRegulations to Support Next Generation Terrestrial Systems andServices Common Law Monitor 2017 httpswwwcommlaw-monitorcom201712articleswireless-2fcc-adopts-a-second-wave-of-millimeter-wave-regulations-to-support-next-genera-tion-terrestrial-systems-and-services
[95] Y Saleem N Crespi M H Rehmani and R Copeland ldquoInter-net of things-aided smart grid technologies architecturesapplications prototypes and future research directionsrdquo IEEEAccess vol 7 pp 62962ndash63003 2019
[96] Y Li X Cheng Y Cao DWang and L Yang ldquoSmart choice forthe smart grid narrowband internet of things (NB-IoT)rdquo IEEEInternet of Things Journal vol 5 no 3 pp 1505ndash1515 2018
[97] P Reininger ldquo3GPP standards for the internet of-thingsrdquohttpswwwslideshareneteikoseidel3gpp-standards-for-the-internetofthings 11 3gpp Standards for IoTpdf 2016
[98] ldquocellular system support for ultra-low complexity and lowthroughput internet of things (CIoT)rdquo httpsportal3gpporgdesktopmodulesSpecificationsSpecificationDetailsaspxspec-ificationId=2719
[99] R Ratasuk B Vejlgaard N Mangalvedhe and A GhoshldquoNB-IoT system for M2M communicationrdquo in Proceedings ofthe IEEE Wireless Communications and Networking Conference(WCNC) pp 1ndash5 2016
[100] Link Labs StaffAnOverview ofNarrowband IoT (NB-IoT) 2018httpswwwlink-labscomblogoverview-of-narrowband-iot
[101] Y E Wang X Lin A Adhikary et al ldquoA primer on 3GPP nar-rowband internet of thingsrdquo IEEE Communications Magazinevol 55 no 3 pp 117ndash123 2017
combizplansm2m-business-plans[104] GSMA Mobile IoT in the 5G Future- NB-IoT and LTE-M
in the context of 5G 2018 httpswwwgsmacomiotwp-contentuploads201805GSMAIoT MobileIoT 5G FutureMay2018pdf
[105] M Contento 5G and IoT ndash Emerging Tech with Endless UseCases 2019 httpswwwtelitcomblogstate-of-5g-and-iot-current-future-applications
[106] The Guardian View on Google Versus Huawei No Winners TheGuardian 2019 httpswwwtheguardiancomcommentisfree2019may20the-guardian-view-on-google-versus-huawei-no-winners