Insert Presenter Photo Here Identifying technology to ... · W-CDMA (R99) EGPRS 4/12 (R99) Growth in peak and average spectral efficiency ... HSDPA cell throughput and geometry factor

Insert Presenter Photo Here

100 x 120 pixel in JPEG or BMP

format

Renaud DuverneWireless R&D

Market Initiative ManagerAgilent Technologies

Identifying technology to deliver the next 100x growth in

Wireless

Page 2

QuizFrom press release announcing WiMAX adopted as IMT-2000 standardhttp://biz.yahoo.com/ap/071019/un_wimax.html?.v=1

“WiMAXTM is capable of delivering wireless broadband connections at speeds of 70 megabits per second or more across an area of up to 40 miles. It's faster than many fixed-line broadband connections today, which typically offer speeds of around 2 megabits per second.”

False

True

100% true but incomplete – The future is all about data densities

Page 3

Peak vs. Average

Page 4

Cooper’s law on wireless capacity growth

Dr. Martin Cooper of Motorola - “father” of the modern mobile phone - has observed:

Dr. Martin Cooper in 1982 with the DynaTAC

The number of simultaneous voice and data connections has doubled every 2.5 years

since wireless began (1900)

Cooper’s Law

Page 5

What is enabling this apparent exponential growth in wireless communications?The capacity of a system to deliver services is defined by three main factors:• The available radio spectrum – in MHz• The efficient use of that spectrum – bits / second / hertz• The number of cells – this equates to spectrum reuse

Number of cells

Effic

ienc

ySpec

trum

Page 6

Growth to date dominated by increasing cell count

If we apply Coopers law over the last 50 years we are looking at a growth in wireless capacity of perhaps 1,000,000Allocating this growth between the axes of capacity looks roughly like this:

Gro

wth

fact

or

1

10

100

1000

20 25

2000

Efficiency Spectrum No. of cells

10000

Growth has historically been dominated by the

increase in the number of cells

Page 7

What is the outlook for growth in the next 10 years?

The bulk of historical connections has been voice, more recently augmented by SMS

To a first approximation Cooper’s law represents growth in wireless capacity

Will the historical growth continue?

What will be the demand?

And more importantly, what will be the supply?

?

Page 8

A singular truth…

Although Leonardo got the limb count wrong (or is this an unfulfilled prediction of MIMO man?) he did observe that for quite some time the average person has just one mouth

On average this mouth gets used for about a couple of hours a month to make cellular phone calls

Page 9

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

CY04 CY05 CY06 CY07 CY08 CY09 CY10

Subs

crib

ers

(M)

Mobile Subscribers by Geographic Region

14.4%390260CALA11.2%40602950ALL

1.4%250240NA10.6%14001030EMEA12.5%20201420APACCAGR20102007Region

Source: Infonetics, June 2007

10%59.7%44.7%Penet.1.2%6.8B6.6BPop.

CAGR20102007World

During 2008 50% of the world’s population will have a mobile phone

APAC - Asia Pacific

EMEA - Europe Middle East Africa

NA – North AmericaCALA – Central / Latin America

Page 10

World population growth

World population

0

1000000

2000000

3000000

4000000

5000000

6000000

7000000

8000000

9000000

10000000

1880 1900 1920 1940 1960 1980 2000 2020 2040 2060

Average growth 3.6% per year from 2005 to 2050

Page 11

Predicting growth in voice traffic

Assuming over the next 10 years:• World population growth remains around 3.6%• Average mouth count remains near 1• Average usage < 3 hours / month• Mobile phone penetration approaches 75%

Conclusion:• Voice traffic growth could double in the next 10 years

• Seems manageable

Page 12

Growth in data traffic

The potential for growth in data is not bounded by physical human attributes like:• Number of people• Number of mouths per person

So what will fuel demand for wireless data?• Without discussing the vexed question of the killer application,

demand will largely be stimulated by the available capacity • Think hard disc drive size!

Page 13

Cellular wireless peak data rates appear to be on track to grow by 100,000 between 1985 and 2015

5608.415 MHz42 MbpsHSDPA+64QAM & 2x2 MIMO2009

2015

2013

2012

2011

2006

2003

2000

1997

1992

1985

Date

108716.3120 MHz326.4 MbpsLTE 4x4 MIMO

5768.6120 MHz172.8 MbpsLTE 2x2 MIMO

1872.815 MHz14 MpbsHSDPA

3335120 MHz100 MbpsLTE

10

0.4

0.2

.07

0.032 - 0.048

0.015

Peak Spectral efficiency

1

1

4 / 12

4 / 12

4 / 12

7 / 21

Frequencyreuse

100 MHz

5 MHz

200 kHz

200 kHz

200 kHz

30 kHz

ChannelBandwidth

1 Gbps

2 Mbps

474 kbps

171 kbps

9.6 – 14.4 kbps

9.6 kbps

Peak data rate

1AMPS

667IMT-Advanced targets

26.6W-CDMA

13.3EDGE

4.7GPRS

2.1 – 3.2GSM

Normalized efficiencySystem

With such peak data rates the demand for capacity could be huge

Page 14

Is wireless keeping up with wired?

Today’s Fiber capacity: 10 Gbps (OC-192)Theoretical capacity of a single fiber > 10 Tbps

To match ONE strand of fibrea wireless solution would need

2 GHz of dedicated spectrum with a spectral efficiency of 5 b/s/Hz

Aug 2007: IEEE decides on next generation Ethernet standard

The Institute of Electrical and Electronics Engineers (IEEE) is to create a single standard that covers both 40gbps and 100gbps Ethernet speeds. The standard should be completed in the next three years.

Page 15

With today’s cellular densities, average data rates are falling behind peak data rates by 10x

Pk Data rates x 2800Efficiency x 40Spectrum x 7

Capacity x 280

10000

100000

1000000

10000000

100000000

1000000000

1985 1990 1995 2000 2005 2010 2015

Peak rates Average Efficiency Spectrum Capacity

The average efficiency, spectrum and capacity plots are normalized

A 10x capacity gap has opened up today!

Only a higher cell density will change this

Page 16

Spectral Efficiency bits / sec / Hz

0.01

0.1

1

10

100

1980 1985 1990 1995 2000 2005 2010 2015Average efficiency Peak efficiency

AMPSGSM

GPRSEDGE

W-CDMA

HSDPA1xEV-DO

LTE802.16e

IS-95C

1xEV-DO(A)EGPRS2 1/3

W-CDMA (R99)EGPRS 4/12 (R99)

Growth in peak and average spectral efficiency

HSDPA (R7)HSDPA (R5)

LTEtarget

EGPRS 1/3 (R99)

Peak efficiency lies around this line

Average efficiency and hence capacitygrowth of deployed systems lags well behind and will level off due to inter-cell interference

Peak efficiency drives up air interface cost &

complexity

You pay for the peak but experience the average

Page 17

Assessing the outlook for capacity growth over the next 10 years• Spectrum• Efficiency• Number of Cells

Page 18

One digital band in 1990 to twenty four in 2008

TDD0240023002400230040TDD0192018801920188039TDD0262025702620257038TDD0193019101930191037TDD0199019301990193036TDD0191018501910185035TDD0202520102025201034TDD0192019001920190033FDD2076875879878814FDD2175674678777713FDD1274672871669812FDD231500.91475.91452.91427.911FDD340217021101770171010FDD601879.91844.91784.91749.99FDD109609259158808FDD5026902620257025007FDD358858758408306FDD208948698498245FDD35521552110175517104FDD2018801805178517103FDD2019901930191018502FDD13021702110198019201

ModeDuplexDownlinkUplinkBand

Lots of bands but not all in the same geography

Page 19

IMT Spectrum

3300 3400 3500 3600 3700 3800 3900 4000 4100 4200 4300 4400 4500 4600 4700 4800 4900 5000

1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900

20252110 2170 26901710

100 500 600 700 800 900 1000200 300 400

5150 470 890 915925960

806450 790698

New for IMT in some countries of

Regions 1 & 3

NewRegion 2

NewGlobal

ExistingIMT

identified

New spectrum is limited. Aggregation of multiple bands adds cost and complexity to the terminals. SDR is not yet the answer.

+

Page 20

Spectrum upside for any one geography

Assume the European model of 340 MHz:• 35+35 MHz of GSM @ 900 MHz• 75 + 75 MHz of GSM @ 1800 MHz• 60 + 60 MHz of UMTS FDD @ 2.1 GHz

Add 70 MHz from UHF bandAdd max 200 MHz from 2.6 GHz bandPlus some 3.5GHz?

Spectrum upside could be 2x

Page 21

Assessing the outlook for capacity growth over the next 10 years• Spectrum• Efficiency• Number of Cells

Page 22

Efficiency is limited by inter-cell interference – the geometry factor

Low SNR High SNR

The geometry factor is the ratio between the serving cell power and interfering cells + noise

At the boundary of any two cells the geometry factor will not exceed -3dB

At the boundary of any three cells the geometry factor will not exceed -4.8 dB

No one has yet invented a spatially aware electromagnetic wave that stops at the desired cell boundary!

Page 23

Geometry factor distribution in urban cells

Geometry factor in dB

Cum

ulat

ive

dist

ribut

ion

0 %

100 %

-30 30

This plot shows the variation in geometry factor across a typical outdoor urban cell

Very high spectral efficiency is only seen when the geometry factor is above 15 dB, which is an environment that 90% of the user population will not experience

In-building penetration loss will degrade performance further

This puts a finite and very low limit on indoor performance when using outdoor transmission systems

90% of users 10% of users

0-20 -10 10 20

Most new high data rate/MIMO

performance targets require geometry

factors experienced by <10% of the user

population

Page 24

HSDPA cell throughput and geometry factor vs. coverage

8

4

-32510152025+

Mbps cellthroughput vs. G factor

>20 dB5%

>15 dB10%

>5 dB50%

>2 dB70%

>-3 dB100%

>10 dB30%

Rel-7 Type 3 receiver (Equalizer plus receive diversity)

Figures derived from typical urban G factor distribution and 3GPP TS 25.101 v7.9.0 Tables 9.8D3, 9.8D4 & 9.8F3 for 3 km/h

Any point on the graph represents the entire cell capacity if all users experience that G factor

The average cell throughput is around 3 Mbps or 0.6 b/s/Hz

Page 25

Average efficiency upside

Remaining gains in efficiency will come from:• Wider channels enabling freq dependent scheduling• MIMO• Beamforming• Interference cancellation• Advanced coding techniques

All high efficiency techniques increase system complexity and costsDriving up today’s best performance - which lies somewhere in the range 0.4 b/s/Hz to 0.8 b/s/Hz will be hard work• Average efficiency is not significantly impacted by increasing the peak efficiency due to its

low-coverage• MIMO gains are one-time only and are dependent on the channel conditions requiring both

high SNR and high multipathHistorical average efficiency has been improving around 3x per decadeA very rough figure for the next decade for affordable average efficiency gains is probably going to be similar to the historical trend at around 3x• Consistent with LTE goals

Page 26

Projecting ahead shows the gap between average and peak rates in loaded cell will grow to 90x

Data rates x 100000Efficiency x 87Spectrum x 13

1100x capacity

A 90x gap will exist by 2015

Again, the only option to increase capacity and average data rates is to increase cell density10000

100000

1000000

10000000

100000000

1000000000

1985 1990 1995 2000 2005 2010 2015

Peak rates Average Efficiency Spectrum Capacity

The average efficiency, spectrum and capacity plots are normalized

Page 27

Assessing the outlook for capacity growth over the next 10 years• Spectrum• Efficiency• Number of Cells

Page 28

Cell number upside

History shows that the bulk of growth in wireless capacity has come from increasing the number of cells Today we are around one cell per 1000 usersThis has huge potential to changeIt is not unreasonable to assume one cell per ten users which could be achieved with deployment of home base stations or femtocells into 30% of householdsFrom the operator’s perspective, growing capacity by having the end user pay for the CapEx and OpEx is very attractive!With this assumption the upside for cell numbers could easily be in the region of 100x

Page 29

Comparing wireless growth potential for the next decade

Gro

wth

pot

entia

l

1

10

32

100

Efficiency Spectrum No. of cells

100 The historical domination of growth delivered by increasing cell numbers will continue into the next decadeThe alternatives of improving average efficiency and adding spectrum look like continuing at their historical levels which is around 15x less significant

Page 30

Which small cell technology will prevail?What about plain old WiFi?Love it or hate it, WiFi is here to stayWiFi today is by far the biggest provider of home and nomadic wireless data services and so can’t be ignored.By cellular standards it is a crude technology

• No power control• No frequency awareness• Limited mobility and handover capability• Limited range

But WiFi’s simplicity and low cost has led to mass deployment

Page 31

The Wild West of municipal WiFi

Network A NetworkB

AP Antenna 7dBi

Site to site pathloss ~60 dB

WiFi ACLR1 ~30 dB

AP TX power 23 dBm

Uplink pathloss ~90 dB

Client power ~15 dBm

Client antenna -5 dBi

RX signal ~ -73 dBm

RX interference ~ -53 dBm

20 dB of interference!Wireless anarchy even on adjacent channels!

Page 32

Who needs adjacent channels anyway?

How can this ever work!!!

Being a collision-based protocol it

sort of works even here

Only WiFi channels 1, 6 and 11 avoid overlap

Eleven APs broadcasting on channel 11!

Page 33

Ten things WiFi has in common with public toilets1. Access is likely to be cheap or increasingly free (beware of 4 star hotels!)2. Usually no need to wait unless you’re in a crowd (E.g. 3GPP meetings!)3. It’s not always available just where you want it and sometimes you just can’t

afford to wait4. The quality is very variable and not regulated - You trust your home, the office,

the VIP lounge, your hotel room - but Grand Central Station?5. Once you find one it may be out of order6. They all have different interfaces7. You might fear you could catch a virus8. Once started, you must finish before moving elsewhere9. You are at the mercy of people who want to look at what you are doing10. If another user is too close you may be splattered by unwanted emissions

But despite their obvious limitations, where would we be without public toilets and WiFi?

Page 34

Often the “best” technology doesn’t win

• Ethernet vs. Token ring• 802.11b vs. HiperLAN• Windows 3.1 vs. Unix• Iridium vs. GSM

And now…• McDonalds vs. McCaw?

Anyone fancy a nybble before dinner?

Would you like bytes with that?

“Perfection is the enemy of the good”

Gustav FlaubertFrench Novelist 1821 - 1880

Page 35

The battle for the voice mass marketIridium vs. GSM

2.5 Billion250,000Subscribers

100x – 1000x ?1Relative Capacity

Scalable / mediumVery highInfrastructure cost

~$0.1 / minute~$1 / minutePrice of usage

Hundreds of thousands66No. of base stations

Urban, some rural100% (Outdoors)Coverage

GSMIridium

Conclusion: Iridium is a niche supplier of a highly valuable serviceGSM owns the voice mass market

Page 36

The battle for the car mass marketTata Nano vs. Bugatti Veyron

And the winner is…

$2,500

(Single band SISO GSM)

$1,500,000

(326.4 Mbps 4x4 MIMO hex-band LTE)

Page 37

The battle for the wireless data mass market4G vs. Hotspot ?

1 Billion?100 Million?Subscribers by 2015

100x ?1Relative Capacity

Scalable / very lowVery highInfrastructure cost

Tending to zeroHighly variablePrice of usage

50 Million?500,000?No. of base stations

HotspotsUrbanCoverage

Hotspot4G

Conclusion: 4G provides highly valuable limited capacity mobile data WiFi owns the high-capacity wireless data market

Page 38

Can femtocells outperform and replace WiFi?

The potential for cellular femtocells to deliver the future growth of wireless is very real but:• the industry remains largely focussed on improving efficiency which

is driving up cost and complexity• many the engineering challenges of femtocells remain to be solved

Page 39

Femtocell key challenges

• It’s all about interference mitigation!• Co-channel deployment looks very problematic except for rural• Adjacent channel seems possible• Regulatory aspects – need GPS for authentication• A hackers paradise – build your own cellular network…• Open vs. closed access• Business models

• Tied to operator

• Net neutrality – who owns the backhaul?• Could blow femtocell competition off the planet – will vary by country

• Could it hurt my cat?• Possible public backlash over radiation concerns?

Page 40

In a mature market, Value > price > cost

€ Y

€ Y

€ Y

€ X

Cost / MByte

€ 0.0023

€ 0.007

€ 0.7 - € 7

€1000

Price / MByte

50 Hours @ 100 kbps

3 GBytes

10 kbps

160 Bytes

Data rate / volume

Unicast Mobile TV (Capped at 50 hrs)

Data service (capped at 3 Gbytes)

Voice

SMS

PriceService

3€20 / month

1€5 / month

300 - 3000€0.05 to €0.5 / minute

400,000€0.15 / message

RatioPrice €

As data rates rise the user value per MByte plummets. Current pricing encourages unsustainable use of highly valuable macro wireless resources

Page 41

Comparison of traditional cellular vs. hotspot for data delivery over next decade

Very lowHigh and not falling fast enoughCost per bit

Low, stableGrowingComplexity

Where its neededReducingCoverage

100x ?6XCapacity

HotspotMacro/Micro

Improvements to the macro network e.g. EDGE Evolution, HSPA+ andLTE need to continue but the bulk of the traffic growth and high data

rates will be delivered by small cell technology

Page 42

So which hotspot technology will win?

The answer lies between the extremes of highly regulated femtocellular or the anarchy of Wild West WiFi:

WiFi enabled iPhoneCellular controlToday the needle can only move

to the left.But how far will it swing?

FemtometerTM

Control Anarchy

?

Page 43

Thank you for listening!