Top Banner
1 Wireless Networks Ivan Marsic Rutgers University
103

1 Wireless Networks Ivan Marsic Rutgers University.

Dec 20, 2015

Download

Documents

Welcome message from author
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
Page 1: 1 Wireless Networks Ivan Marsic Rutgers University.

1

Wireless Networks

Ivan Marsic

Rutgers University

Page 2: 1 Wireless Networks Ivan Marsic Rutgers University.

2

ISO OSI Protocol Stack

• Protocol at layer i doesn’t know about protocols at i1 and i1

1: Physical

2: LinkMAC

3: Network

4: Transport

• Radio spectrum• Infrared• Fiber• Copper

• IEEE 802.11 WiFi• IEEE 802.3 Ethernet• PPP (modems, T1)

• End-to-end (IP)• Routing• Address resolution

• Reliable (TCP)• Unreliable (UDP)

Application

Page 3: 1 Wireless Networks Ivan Marsic Rutgers University.

3

Infrastructure vs. Ad Hoc (1)

(a) (b)

Internet/WebInternet/Web

802.11

802.3

Mobile Node

Access Point

Mobile Node

Internet/WebInternet/Web

802.11

802.3

Mobile Node

Access Point

Mobile Node

802.11

Mobile Node

Mobile Node

Mobile Node

Mobile Node

802.11

Mobile Node

Mobile Node

Mobile Node

Mobile Node

Page 4: 1 Wireless Networks Ivan Marsic Rutgers University.

4

Infrastructure vs. Ad Hoc (2)

(c) (d)

Page 5: 1 Wireless Networks Ivan Marsic Rutgers University.

5

Waves

Page 6: 1 Wireless Networks Ivan Marsic Rutgers University.

6

EM Spectrum Allocation

Freq.

Freq.

Freq.902 MHz 928 MHz 2.4 GHz 2.4835 GHz 5.725 GHz 5.785 GHz

(AM radio)MF

(SW radio)HF

(FM radio - TV)VHF

(TV – Cell.)UHF SHFLF

30 KHz 300 KHz 3 MHz 30 MHz 300 MHz 3 GHz 30 GHz

Vis

ible

Infr

are

d

X raysUV

Gammarays

1 KHz 1 MHz 1 GHz 1 THz 1 PHz 1 EHz

IEEE 802.11aHiperLAN II

IEEE 802.11b, gBluetooth

Microwave ovens

Cordless phonesBaby monitors

(old) Wireless LANs

ISM UNII

Page 7: 1 Wireless Networks Ivan Marsic Rutgers University.

7

Communication Process

Information Source

Information Source TransmitterTransmitter ReceiverReceiver

NoiseSource

NoiseSource

DestinationDestinationCommunication

Channel

0 1 0 0 1 1 0 0 1 0

InputSignal:

SourceData:

Noise:

OutputSignal:

SamplingTimes:

0 1 0 0 1 1 0 0 1 0SourceData:

1 1 0 0 1 1 1 0 1 0DataReceived:

Bits in error

S

N

S+N

Decisionthreshold

Page 8: 1 Wireless Networks Ivan Marsic Rutgers University.

8

Decibel definition

Link, channel,

repeater, or node

Power in Power out

Page 9: 1 Wireless Networks Ivan Marsic Rutgers University.

9

Fourier Series Approximation

0

30

60 360180

90

120

150 210

240

270

300

330

30

6090

Angular phase of fundamental wave

Time

Resultant

Dis

pla

cem

en

t

Three dots runningat the same speedaround the circle indifferent lanes:

Page 10: 1 Wireless Networks Ivan Marsic Rutgers University.

10

Phase Space

Time (t)

Am

plit

ud

e Period (T)

A

Phase ()

Frequency = 2 / T

A sin ( t )

Phase space:

A

02

/ 2

3 / 2

[rad / s]

Page 11: 1 Wireless Networks Ivan Marsic Rutgers University.

11

Wireless Transmission and Receiving System

Modulator

Destination

Error Control Encoder

Source Decoder (Decompress)

Source Encoder (Compress)

Error Control Decoder

InformationSource

Demodulator

Tra

nsm

itte

r

Re

ceiv

er

CommunicationChannel

Page 12: 1 Wireless Networks Ivan Marsic Rutgers University.

12

Modulation

Carrier signal

Analog message signal

Amplitude-modulated (AM) signal

Frequency-modulated (FM) signal

Phase-modulated (PM) signal

Carrier signal

Digital message signal

Frequency-shift-keying (FSK) signal

Amplitude-shift-keying (ASK) signal

Phase-shift-keying (PSK) signal

1 0 1 1 0

Page 13: 1 Wireless Networks Ivan Marsic Rutgers University.

13

Modulation—PSK

135

225315

45

0001

10 11

90

180

2700 00

01

10

11

270

000

010

100

110

315225

180

135 90

45

111

001011

101

(a) (b) (c)

0000

0010

0100

0110

1000

1010

1100

1110

0001

0011

0101

0111

1001

1011

1101

1111

Page 14: 1 Wireless Networks Ivan Marsic Rutgers University.

14

Example 2.1

3 bits

110

3 bits

100

3 bits

010Amplitude

Time

Page 15: 1 Wireless Networks Ivan Marsic Rutgers University.

15

Gaussian r.v. and Q-function

x

p(x)

(2)

1

0

x

½

1

F(x)

0

(a) (b)

x

p(x)

0 z

Pr{x z}

(c) (d) 1 2010 6

10 5

10 4

10 3

10 2

10 1

10 0

3 4

z

Q(z)

Page 16: 1 Wireless Networks Ivan Marsic Rutgers University.

16

Effect of Noise on Signal

Decisionthreshold

0

VA

VB

Symbol A

Symbol B

ModulatorModulator DemodulatorDemodulatorbits (symbols) in bits (symbols) out

Noise ~ N(0,)Noise ~ N(0,)

Bit-to-waveform mapper

Decision device

++

(a)

(b)

Noi

se p

df

P(AR|BT)

P(BR|AT)

Channel

Page 17: 1 Wireless Networks Ivan Marsic Rutgers University.

17

Probability of Error for2,4-PSK

2 4010 7

10 6

10 5

10 4

10 3

10 2

10 1

6 8

SNR per bit, Eb/N0 (dB)

Pro

ba

bil

ity

of

bit

err

or,

Pe

10

BPSK

QPSK

12 14

Page 18: 1 Wireless Networks Ivan Marsic Rutgers University.

18

Discrete vs. Continuous Channel

(b) (c)

0 1 0 0 1 1 0 0 1 0

InputSignal:

SourceData:

Noise:

OutputSignal:

(a)

(1,0,1) (1,1,1)

(0,0,1)

(1,0,0)

Page 19: 1 Wireless Networks Ivan Marsic Rutgers University.

19

Signals as Vectors

s(t)

t5 V

T

Example 3-bit message: 1 0 1

A three-bit signal waveform

0 T 2T 3T

p1(t)

p2(t)

p3(t)

Orthogonal function set (Basis vectors)

s1(t)

s2(t)

s3(t)

(0,0,0)

(1,1,1) (1,1,1)

(1,1,1)

(1,1,1) (1,1,1)

(1,1,1)

(1,1,1)

(b)

(a)

(c)

(d)

Page 20: 1 Wireless Networks Ivan Marsic Rutgers University.

20

11S

N 1

1S

N rS = 2

1,1 1,1

S

rN= 1/2 1, 1 1, 1

N

rS = 2

1,1 1,1

S

rN= 1/2 1, 1 1, 1

N

1, 1,1

1, 1,1

1, 1, 1

1, 1, 1

1,1, 1

1,1, 1

111

S

N

1,1,1

1, 1,1

1, 1,1

1, 1, 1

1, 1, 1

1,1, 1

1,1, 1

111

S

N

1,1,1

Geometric Representation

(c)(a) (b)

Page 21: 1 Wireless Networks Ivan Marsic Rutgers University.

21

Signal Space

ST

N

SR)(2 NSBT

(b)(a)

Page 22: 1 Wireless Networks Ivan Marsic Rutgers University.

22

Locus of Error-Causing Signals

(b)(a)

h

O'N

ST

Noise sphere,radius ,centered on ST

O

Signal sphere,radius

SR = ST + N

BTS2

BTN22h

O'

2h

O'

Page 23: 1 Wireless Networks Ivan Marsic Rutgers University.

23

Error Detection and Correction

(a) (b)

1,1 1,1

1,1 1,1Validmessage

Validmessage

Errormessage

Errormessage

r2 = 22r1 = 2

r3 = 23

1, 1, 1

1,1,11, 1,1

1, 1,1

1,1, 1

111

r2 = 22r1 = 2

r3 = 23

1, 1, 1

1,1,11, 1,1

1, 1,1

1,1, 1

111

Page 24: 1 Wireless Networks Ivan Marsic Rutgers University.

24

Wave Interactions

(a)

(b)

Rough ceiling

Transmitter

“Knife-edge”obstacle

Receiver 1

s

s1

s0 + s1' + s2'

Receiver 2

s1'' + s2''

s0

s2

Rough ceiling

Transmitter

“Knife-edge”obstacle

Receiver 1

s

s1

s0 + s1' + s2'

Receiver 2

s1'' + s2''

s0

s2

Page 25: 1 Wireless Networks Ivan Marsic Rutgers University.

25

Interference & Doppler Effect

(a) (b)

s0

s1

s2 s0+s1+s2

s0

s1

s2 s0+s1+s2vv

Page 26: 1 Wireless Networks Ivan Marsic Rutgers University.

26

Multipath Fading (1)

Transmittedsignal

Transmittedsignal

Rec

eiv

edLO

S s

igna

l

Rec

eiv

edLO

S s

igna

l

ReceivedNLOS signals

ReceivedNLOS signals

Time

Time

Page 27: 1 Wireless Networks Ivan Marsic Rutgers University.

27

(b)(a)

Plane-Earth Model

Receiver

Source

Receiver

Source

Receiver

Source

Receiver

Source

Page 28: 1 Wireless Networks Ivan Marsic Rutgers University.

28

Delay Spread

(a) (b)

Tx Rx

2

1

Tx Rx

2

1

Delay (s)

Pat

h po

wer

(dB

)

0 1 2 3 4 5 6

mxd

Delay (s)

Pat

h po

wer

(dB

)

0 1 2 3 4 5 6

mxd

Page 29: 1 Wireless Networks Ivan Marsic Rutgers University.

29

Discrete-time Delay Model

g(t, 0) g(t, ) g(t, 2) g(t, 3) g(t, N)

r(t)

š(t)

g(t, 0) g(t, ) g(t, 2) g(t, 3) g(t, N)

r(t)

š(t)

Page 30: 1 Wireless Networks Ivan Marsic Rutgers University.

30

Multipath Fading (2)

Direct path (1 component)

Delay spread (2 components)Flat Fading

Delay spread (2 components)Frequency Selective Fading

Doppler spread (2 components)Fast Fading

0

0.05

0.05

0.1

0.1

0.5 1.0 1.5 2.0 2.5 3.00

0.15

0

0.05

0.05

0.1

0.1

0.5 1.0 1.5 2.0 2.5 3.00

0.15

0

0.05

0.05

0.1

0.1

0.15

3.00 0.5 1.0 1.5 2.0 2.5

0.2

0

0.05

0.05

0.1

0.1

0.15

3.00 0.5 1.0 1.5 2.0 2.5

0.2

0.5 1.0 1.5 2.0 2.5 3.00

0

0.05

0.05

0.1

0.5 1.0 1.5 2.0 2.5 3.00

0

0.05

0.05

0.1

0.05

0.05

0.2

0.5 1.0 1.5 2.0 2.5

0.15

3.00

0

0.1

0.05

0.05

0.2

0.5 1.0 1.5 2.0 2.5

0.15

3.00

0

0.1

Page 31: 1 Wireless Networks Ivan Marsic Rutgers University.

31

Error Probabilities

2 40 6 8

SNR per bit (dB)

Pro

ba

bil

ity

of

bit

err

or,

Pe

10

BPSKAWGN

12 1410 5

10 4

10 3

10 2

10 1

10 0

BPSKRayleigh

16 18 20

Page 32: 1 Wireless Networks Ivan Marsic Rutgers University.

32

Medium Access Control (MAC)

• Controls who gets to transmit when• Avoids “collisions” of packet

transmissions

Station 1

Station m

1

2

Station 2

m

Packetarrival rate Receiver

Page 33: 1 Wireless Networks Ivan Marsic Rutgers University.

33

Coordination Problem

Page 34: 1 Wireless Networks Ivan Marsic Rutgers University.

34

Collisions

Receiver electronics detects collision Receiver broadcasts info about collision (jam)

= total time to detect collision = RTT of the most distant station

ReceiverReceiver

Station 1

Station m

Station 2 Station 2

ReceiverReceiver

Station 1

Station m

Station 2 Station 2

Page 35: 1 Wireless Networks Ivan Marsic Rutgers University.

35

Channel State

Assumption:There is always at least one station in need of transmission

Objective:Maximize the fraction of time for the “Successful transmission” state( or: minimize the duration of “Idle” and “Collision” )

Time

SuccessfultransmissionIdle

Error/Collision

Page 36: 1 Wireless Networks Ivan Marsic Rutgers University.

36

Multiaccess vs. Multiplexing

MUX

Ordering of packets on higher capacity link

Receiver

Ordering of packets on shared medium

Page 37: 1 Wireless Networks Ivan Marsic Rutgers University.

37

Static multiaccess schemes: TDMA and FDMA

Deterministic Schemes

m1 3

time

2 m1 32 m1 32

TDMA

freq

uenc

y

m1 3

time

2 m1 32 m1 32

TDMA

freq

uenc

y

m

321

time

m

321

m

321

FDMA

freq

uenc

y

m

321

time

m

321

m

321

FDMA

freq

uenc

y

Page 38: 1 Wireless Networks Ivan Marsic Rutgers University.

38

Poisson Arrivals Model

0

5

10

15

20

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

arrivals per time unit (n)

per

cen

t o

f o

ccu

rren

ces

(%)

Page 39: 1 Wireless Networks Ivan Marsic Rutgers University.

39

Parameter

t

txmit

Transmitter

Receiver

< 1

t

txmit

» 1

L

C

txmit

Propagation constant :

• Ratio of propagation delay vs. packet transmission time

Page 40: 1 Wireless Networks Ivan Marsic Rutgers University.

40

Vulnerable Period

• Packet will not suffer collision if no other packets are sent within one packet time of its start

tstart

tstart txmit

Vulnerable periodtstart txmit

Current packetCollides with

the head of thecurrent packet

Collides withthe tail of the

current packet

Time

Page 41: 1 Wireless Networks Ivan Marsic Rutgers University.

41

ALOHA Protocols

TimePacketArrivals

Departures

1 2 3 4 5 6 7

1 2 3 4 5 6 7

Time

Departures

1 2 3 4 5 6 7

1 2 3 4 5

ALOHA

SlottedALOHA

PacketArrivals

Page 42: 1 Wireless Networks Ivan Marsic Rutgers University.

42

Transmission Success Rate

Time

Departures Slot 1

Arrivals atStation 1

Slot 2 Slot 3 Slot 4

Departures Slot 1

Arrivals atStation k

Slot 2 Slot 3 Slot 4

Time

Time

(a) – Slotted ALOHA (b) – Pure ALOHA

PacketTime

Time

Receiver

1

k

Page 43: 1 Wireless Networks Ivan Marsic Rutgers University.

43

ALOHA and Slotted ALOHA State Diagram

SendWait for

start of slot

ALOHA

Abort

ACK arrived /

Collision /

End

Backoff CWmax /Backoff CWmax /

S-ALOHA

Wait for backoffS-ALOHA

Wait for ACK

Increase backoff

ALOHA

SendWait for

start of slot

ALOHA

Abort

ACK arrived /

Collision /

End

Backoff CWmax /Backoff CWmax /

S-ALOHA

Wait for backoffS-ALOHA

Wait for ACK

Increase backoff

ALOHA

Page 44: 1 Wireless Networks Ivan Marsic Rutgers University.

44

Analysis of Slotted ALOHA (1)

ASSUMPTIONS FOR ANALYSIS:• All packets require 1 slot for x-mit• Poisson arrivals, arrival rate • Collision or perfect reception (no errors)• Immediate feedback (0, 1, e)• Retransmission of collisions

(backlogged stations)• No buffering or infinite set of stations

(m = )i 1 i i 1 i 2

Time Slots

Page 45: 1 Wireless Networks Ivan Marsic Rutgers University.

45

Backlogged Stations

• “Fresh” stations transmit new packets

• “Backlogged” stations re-transmit collided packets

ReceiverReceiver

Fresh Station

Backlogged Station

/m

/m

“Fresh” station

“Backlogged” station

“Fresh” station

“Backlogged” station

Page 46: 1 Wireless Networks Ivan Marsic Rutgers University.

46

ALOHA System Model (1)

• In equilibrium state, system input equals system output

= S = GeG

SystemReceiverReceiver

/m

/m/m

/m

System Input =

System Output = S

Transmission Attempts = G

“Fresh” station

“Backlogged” station

Page 47: 1 Wireless Networks Ivan Marsic Rutgers University.

47

ALOHA System Model (2)

Channel

Stations Receiver

G S

ChannelG

SGP0

G(1 P0)

Page 48: 1 Wireless Networks Ivan Marsic Rutgers University.

48

Analysis of Slotted ALOHA (2)

• 0 < < 1, since at most 1 packet / slot

• Equilibrium: departure rate = arrival rate

• Backlogged stations transmit randomly

• Retransmissions + new transmissions:Poisson process with parameter G >

• The probability of successful x-mit: S=GP0,where P0=prob. packet avoids collision

• No collision => no other packets in the same slot: GGetAtAPGGPS 0)()1(0

Page 49: 1 Wireless Networks Ivan Marsic Rutgers University.

49

Efficiency of ALOHA’s

S-ALOHA: In equilibrium, arrival rate = departure rate: = GeG

Max departure rate (throughput) = 1/e 0.368 @ G = 1

0 0.5 1.0 1.5 2.0 3.0

0.1

0.2

0.3

0.4Slotted ALOHA: S = Ge–G

Pure ALOHA: S = Ge–2G

G (transmission attempts per packet time)

Arrival rate

Equilibrium

S(t

hrou

ghpu

t p

er p

acke

t tim

e)

Page 50: 1 Wireless Networks Ivan Marsic Rutgers University.

50

Unslotted (Pure) ALOHA

• Assume: all packets same size, but no fixed slots

• The packet suffers no collision if no other packet is sent within 2 packets long: S=GP0=Ge2G

• Max throughput 1/2e 0.184 @ G = 0.5• Less efficient than S-ALOHA, but

simpler, no global time synchronization

i

Page 51: 1 Wireless Networks Ivan Marsic Rutgers University.

51

Markov chain for S-ALOHA

P10

0 1 2 3 4

P21

P12

P00 P11 P22

P02

P03P04

P33 P44

P32

P23

Page 52: 1 Wireless Networks Ivan Marsic Rutgers University.

52

Instability of Slotted ALOHA

Desiredstablepoint

Departure rateGe G

Unstableequilibrium

Undesiredstablepoint

mqa

G = (m n)qa nqr

G = mqrG = mqa

Arrival rate(m n)qa

n = mn = 0

G = 0

S(t

hrou

ghpu

t per

slo

t)

G (transmission attempts per slot)

Negative drift Positive drift

Page 53: 1 Wireless Networks Ivan Marsic Rutgers University.

53

Carrier Sensing (CSMA)

• Listen before talk (unlike ALOHA, where talk when you need to)

Page 54: 1 Wireless Networks Ivan Marsic Rutgers University.

54

CSMA/CD

1. Wait until the channel is idle.

2. When the channel is idle, transmit immediately and listen while transmitting.

3. In case of a collision, stop the packet transmission, and then wait for a random delay and go to step 1.

IEEE 802.3 (Ethernet)

Page 55: 1 Wireless Networks Ivan Marsic Rutgers University.

55

Basic CSMA Protocols

CSMA Protocol

Transmission Rules

Nonpersistent If medium is idle, transmit.If medium is busy, wait random amount of time and sense channel again.

1-persistent If medium is idle, transmit.If medium is busy, continue sensing until channel is idle;then transmit immediately.

p-persistent If medium is idle, transmit with probability p.If medium is busy, continue sensing until channel is idle;then transmit with probability p.

Page 56: 1 Wireless Networks Ivan Marsic Rutgers University.

56

CSMA Protocols State Diagram

Sense Send End Increase backoff

Wait random

time

Wait for IFS

Wait 1 slot

Idle (p-persistent)Pr(1-p) / Idle (p-persistent)

Pr(p) /

No Collision(CSMA/CD) /

ACK received(CSMA/CA) /

Abort

Wait for backoff

Backofftoo large /

Jam

Idle(CSMA/CA) /

Busy(nonpersistent) /

Busy (1-persistent& p-persistent) /

Idle (1-persistent& nonpersistent) /

Timeout(CSMA/CA) /

Collision(CSMA/CD) /

Backoff OK /

Sense Send End Increase backoff

Wait random

time

Wait for IFS

Wait 1 slot

Idle (p-persistent)Pr(1-p) / Idle (p-persistent)

Pr(p) /

No Collision(CSMA/CD) /

ACK received(CSMA/CA) /

Abort

Wait for backoff

Backofftoo large /

Jam

Idle(CSMA/CA) /

Busy(nonpersistent) /

Busy (1-persistent& p-persistent) /

Idle (1-persistent& nonpersistent) /

Timeout(CSMA/CA) /

Collision(CSMA/CD) /

Backoff OK /

Page 57: 1 Wireless Networks Ivan Marsic Rutgers University.

57

Nonpersistent CSMA

Station 1

Time

Station 2

Station 3

Station m

Vulnerableperiod

Packet A Packet B

Packet C

Idleperiod

Vulnerableperiod

1 1

1

0 1 1

Successful transmission(period = 1)

Collision(period = 1+Y)

Y

Page 58: 1 Wireless Networks Ivan Marsic Rutgers University.

58

Efficiency of CSMA protocols

(a)

(b)SlottedALOHA

Pure ALOHA

1-persistent CSMA

NonpersistentCSMA

NonpersistentCSMA/CD

S(t

hrou

ghpu

t per

pac

ket t

ime)

G (transmission attempts per packet time)0.1

0.2

0.4

0.6

0.8

1.0

1 10 100 1000

= 0.01

SlottedALOHA

Pure ALOHA

1-persistent CSMA

NonpersistentCSMA

NonpersistentCSMA/CD

S(t

hrou

ghpu

t per

pac

ket t

ime)

G (transmission attempts per packet time)0.1

0.2

0.4

0.6

0.8

1.0

1 10 100 1000

= 0.01

g = (2)

g (transmission attempts per slot time)

1 2

Arrival rate

Equilibrium

Departure rate

1 ge g

1 e g

Thr

ough

put p

er s

lot

time

g = (2)

g (transmission attempts per slot time)

1 2

Arrival rate

Equilibrium

Departure rate

1 ge g

1 e g

Thr

ough

put p

er s

lot

time

Page 59: 1 Wireless Networks Ivan Marsic Rutgers University.

59

Delay vs. Arrival Rate

Arrival rate

Ave

rage

pac

ket

dela

y

ALOHA

CSMA

TDMA

Max

imum

cha

nnel

tran

smis

sion

rat

e

Page 60: 1 Wireless Networks Ivan Marsic Rutgers University.

60

Hidden and Exposed Terminals

Range of B’s transmissionsRange of A’stransmissions

Hidden Terminal Exposed Terminal

CAB

CAB

AB

CD

AB

CD

Page 61: 1 Wireless Networks Ivan Marsic Rutgers University.

61

CSMA/Basic Atomic Exchange

Data Packet

Receiver

Sender

Covered Station

Busy

Busy

Busy

Hidden Station

Busy

Access to medium deferred

Time

IFS(2)

Idle contentionperiod

Vulnerableperiod =

Vulnerable period = Packet time

Page 62: 1 Wireless Networks Ivan Marsic Rutgers University.

62

CSMA/MACA Atomic Exchange

RTS Data Packet

CTS

Access to medium deferred

Receiver

Sender

Covered Station

Busy

Busy

Busy

Hidden Station

Busy

Access to medium deferred

Time

IFS(2)

Idle contentionperiod

IFS IFS

Vulnerableperiod =

Vulnerable period= RTS + IFS +

Page 63: 1 Wireless Networks Ivan Marsic Rutgers University.

63

RTS/CTS Exchange (1)

CAB

RTS(N-bytes)

Page 64: 1 Wireless Networks Ivan Marsic Rutgers University.

64

RTS/CTS Exchange (2)

CAB

CTS(N-bytes)

Page 65: 1 Wireless Networks Ivan Marsic Rutgers University.

65

RTS/CTS Exchange (3)

CAB

N-bytes Packet

Defer(N-bytes)

Page 66: 1 Wireless Networks Ivan Marsic Rutgers University.

66

Components of 802.11 LANs

Ad hoc network does not have distribution system nor access point

Distribution system

Station

Wireless medium

Access point

Page 67: 1 Wireless Networks Ivan Marsic Rutgers University.

67

IBSS and Infrastructure BSS

Independent BSS (IBSS) Infrastructure BSS

Access point

Page 68: 1 Wireless Networks Ivan Marsic Rutgers University.

68

Extended Service Set (ESS)

DS

t = 1 t = 2

BSS1 BSS2 BSS3

AP3AP2AP1

Page 69: 1 Wireless Networks Ivan Marsic Rutgers University.

69

802.11 Network Services

Service Provider Description

Distribution Distribution Service used by stations to exchange MAC frames when the frame must traverse the DS to get from a station in one BSS to a station in another BSS.

Integration Distribution Frame delivery to an IEEE 802 LAN outside the wireless network.

Association Distribution Used to establish a logical connection between a mobile station and an AP. This connection is necessary in order for the DS to know where and how to deliver data to the mobile station.

Reassociation Distribution Enables an established association to be transferred from one AP to another, allowing a mobile station to move from one BSS to another.

Disassociation Distribution Removes the wireless station from the network.

Authentication Station Establishes identity prior to establishing association.

Deauthentication

Station Used to terminate authentication, and by extension, association.

Privacy Station Provides protection against eavesdropping.

MSDU delivery Station Delivers data to the recipient.

Page 70: 1 Wireless Networks Ivan Marsic Rutgers University.

70

States and Services

State 1:Unauthenticated,

Unassociated

Class 1Frames

SuccessfulAuthentication

State 2:Authenticated,Unassociated

State 3:Authenticated

and Associated

De-authenticationNotification Disassociation

Notification

SuccessfulAuthentication orRe-association

De-authenticationNotification

Class 1 & 2Frames

Class 1, 2 & 3Frames

Page 71: 1 Wireless Networks Ivan Marsic Rutgers University.

71

802.11 Interframe Spacings

Busy Frame transmission.....

Contention period

Backoff slots

DIFS

PIFS

SIFS

Time

Defer access Select slot using binary exponential backoff

Page 72: 1 Wireless Networks Ivan Marsic Rutgers University.

72

Basic 802.11 Transmission Mode

TimeDIFS

Receiver

Sender

Data

SIF

S

ACK

4 3 2 1 0

BackoffBusyBusy

BusyBusy

Page 73: 1 Wireless Networks Ivan Marsic Rutgers University.

73

802.11 Protocol State Diagram – Sender

Sense Send End

Increase CW & Retry count

Wait for end of transmission

Wait for DIFS

ACK error-free /

Abort

Wait for IFS

Sense

SenseCountdown for backoff

while medium idle

Busy /

Idle /

Busy /

Idle /

Busy /

Backoff == 0 /

Timeout /

Backoff 0 /

Retry Retrymax /

New

pac

ket

/

Retry Retrymax /

Wait for EIFS

1

SetBackoff

Idle /

1

1

ACK in error /

Page 74: 1 Wireless Networks Ivan Marsic Rutgers University.

74

802.11 Protocol State Diagram – Receiver

ReceiveSend ACK

EndWait for

SIFS

Wait for EIFS

Packet in error /

Packet error-free /

Page 75: 1 Wireless Networks Ivan Marsic Rutgers University.

75

Example: Infra BSS

Assume Station A has a single packet to transmit to B

A B

Accesspoint

DIFSStation A

AP

Station B

DIFS Data SIF

S

ACK

Example: backoff = 4

4,3,2,1,0

Data

SIF

S

ACK

Time

Page 76: 1 Wireless Networks Ivan Marsic Rutgers University.

76

Timing Diagrams

Timing of successful frame transmissions under the DCF.

Frame retransmission due to ACK failure.

Frame retransmission due to an erroneous data frame reception.

Frame

DIFS

SIFSACK

FrameDIFS

Backoff

SIFSACK

Busy FrameDIFS

Backoff

SIFSACK

FrameEIFS

Backoff

SIFSACK

Busy FrameDIFS

Backoff FrameACK Timeout

Backoff

SIFSACK

(a) Time

(b)

(c)

Packet arrival,channel idle

No backoff

Page 77: 1 Wireless Networks Ivan Marsic Rutgers University.

77

Backoff Mechanism

The backoff mechanism of 802.11 MAC.The Frame* transmission time includes the RTS/CTS exchange and the MAC layer ACK.CP: Contention period.

DIF

S

Frame*

Frame*CP

CP CPCP

STA 1

STA 2

STA 3

5 4 3 2 1 0

2 1 0

9 8 7 6 5 4 3 2 1 0

Frame*

3 2 1 0

7 6 5 4 3 2 1 0

4 3 2 1 0

4 3 2 1 0

3 2 1 0

DIF

S

DIF

S

DIF

S

DIF

S

7 6 5 4 3 2 1 0

Frame*

Remainder Backoff

1 0

Frame*

4 3 2 1 0

4 3 2 1 0

Page 78: 1 Wireless Networks Ivan Marsic Rutgers University.

78

RTS/CTS Transmission Mode

RTS

SIF

S

TimeData

CTS

SIF

S

SIF

S

ACK

DIFS

Access to medium deferred

Receiver

Sender 4 3 2 1 0

Backoff

DIFS

Covered Station

BusyBusy

BusyBusy

BusyBusyNAV (RTS)

NAV (CTS)

NAV (Data)

8 7 6

Backoff

5 4

DIFS

Hidden Station

BusyBusyBackoff

Access to medium deferred

Page 79: 1 Wireless Networks Ivan Marsic Rutgers University.

79

802.11 MAC Frame Format

FC D/I Address Address Address SC Address Frame body FCS FC D/I Address Address Address SC Address Frame body FCS

Protocol Type Subtype To From MF RT PM MD W Oversion DS DS

Protocol Type Subtype To From MF RT PM MD W Oversion DS DS

bits 2 2 4 1 1 1 1 1 1 1 1

bytes 2 2 6 6 6 2 6 0 to 2312 4

FC = Frame controlD/I = Duration/Connection IDSC = Sequence controlFCS = Frame check sequence

DS = Distribution systemMF = More fragmentsRT = RetryPM = Power management

MD = More dataW = Wired equivalent privacy (WEP) bitO = Order

Page 80: 1 Wireless Networks Ivan Marsic Rutgers University.

80

802.11 Performance Analysis

Appl

STA 1

Appl

2

Appl

m

Channel

cr r r

Page 81: 1 Wireless Networks Ivan Marsic Rutgers University.

81

3 2 1 03 2 1 07 6 5 4 37 6 5 4 35 4 3 2 1 05 4 3 2 1 0

Idle

DIFSBckof

Inter-eventeligible period 2b

Event 4: Packet 2transmission

RTSBackoff

Idle

Inter-eventeligible period 3

Event 5: Packet 2re-transmission

Bsy

DIFS

Cha

nnel

stat

e

Event 1:Packet 1 arrival

IdleBusyBusy Busy

DIFSBackoff

Inter-eventeligible period 1

RTS

SIF

S

CTS

Data

SIF

S

SIF

S

ACK

DIFS

Idle

Inter-eventeligible period 2a

Event 2: Packet 1transmission

Another stationtransmits

Event 3:Packet 2 arrival

Busy

Time

RTS

CTSSIF

S

SIF

S

BusyBusy

DataDataBackoff

4 3 2 1 04 3 2 1 0

Page 82: 1 Wireless Networks Ivan Marsic Rutgers University.

82

Transmission Example

Busy

6

Busy/success

Idle

DIFS

Actual channel stateobserved by AP

12

transmitattempt

Idle BusyBusy

789101112

Channel stateobserved by A

RTSSIFS SIFS

DataSIFS

ACK

DIFS

CTS Idle

Busy

RTS RTS

DIFS

23456

Idle

Idle

Busy/collision

Page 83: 1 Wireless Networks Ivan Marsic Rutgers University.

83

0CW(0)

K = 1C = 0

K = 0C = 0

K = 5C = 0

K = C = 0

K = 1C = 0

01

11

51

1

pbusy

pidle

pbusy

pidle

pbusy

pidle

pbusy

pidle

pidle pidle

pbusy

1CW(1)

pidle pidle

pbusy

5CW(5)

pidle pidle

pbusy

CW(5)

pidle pidle

pbusy

CW(1)

pcoll~pcoll~

CW(1)

pcoll~pcoll~

CW(1)

pcoll~pcoll~

CW(2)

pcoll~pcoll~

CW(2)

pcoll~pcoll~

CW(5)

pcoll~pcoll~

CW(5)

pcoll~pcoll~

CW(5)

pcoll~pcoll~

CW(5)

pcoll~pcoll~

CW(5)

pcoll~pcoll~

psuccpidle~psuccpidle~ psuccpidle

~psuccpidle~

psuccpbusy/CW(0)~psuccpbusy/CW(0)~

(psuccpbusy pcoll)/CW(0)~ ~(psuccpbusy pcoll)/CW(0)~ ~

psuccpbusy/CW(0)~psuccpbusy/CW(0)~

CW(5)

pcoll~pcoll~

pbusy/CW(0)

pidle

psucc~psucc~

1

Page 84: 1 Wireless Networks Ivan Marsic Rutgers University.

84

802.11 Protocol Architecture

802.11 MAC

802.11bDSSS

802.11aOFDM

802.11DSSS

802.11FHSS

Page 85: 1 Wireless Networks Ivan Marsic Rutgers University.

85

802.11 PHY Frame Using DSSS

PLCP headerPLCP preamble

Payload(variable)

Synchronization (128bits)

SFD (16 bits)

Signal (8 bits)

Service (8 bits)

Length (16 bits)

HEC (16 bits)

Page 86: 1 Wireless Networks Ivan Marsic Rutgers University.

86

IEEE 802.11b BER vs. SNR

SNR (dB) 5 0 5 10 10

10 6

10 5

10 4

10 3

10 2

10 1

BE

R

DBPSK

DQPSK

CCK 5.5 CCK 11

Page 87: 1 Wireless Networks Ivan Marsic Rutgers University.

87

IEEE 802.11b Throughput vs. SNR

1

2

3

4

5

6

7

0

SNR (dB)

Thr

ough

put

(Mb

ps)

DBPSK

DQPSK

CCK 5.5

CCK 11

0 5 10 15 5 20

Page 88: 1 Wireless Networks Ivan Marsic Rutgers University.

88

11Mbps

5.5Mbps

2Mbps

1MbpsDBPSK

DQPSK

DQPSK/CCK

DB

PS

K/C

CK

160 m270 m400 m

550 m

W-LAN Transmission Rates

11 Mbps 8 % of coverage area 1 Mbps 47 % of coverage area Low probability of having good link!!Lucent ORiNICO 802.11b

outdoors, no obstruction—ideal conditions!

Access Point

Mobile Node

Obstacle

Page 89: 1 Wireless Networks Ivan Marsic Rutgers University.

89

Asymmetry

B

A

Access Point Hearing Range

Access Point Transmission Range

AP1

3

Page 90: 1 Wireless Networks Ivan Marsic Rutgers University.

90

Receiver-Based Autorate MAC Protocol

NAV updated using ratespecified in the data packet

AB

CD

1: RTS2: CTS

3: Data

CTS at 1 Mbps

RTS at 2 Mbps

Data at 1 Mbps

Page 91: 1 Wireless Networks Ivan Marsic Rutgers University.

91

IEEE 802.11b Channels

NOTE: The 12 channels in 802.11a do NOT overlap

1:2.

412

2:2.

417

3:2.

422

4:2.

427

5:2.

432

6:2.

437

7:2.

442

8:2.

447

9:2.

452

10:

2.45

7

11:

2.46

2

22 MHz

5 MHz

2.4 GHz 2.483 GHz

Page 92: 1 Wireless Networks Ivan Marsic Rutgers University.

92

Power Conservation

Page 93: 1 Wireless Networks Ivan Marsic Rutgers University.

93

Comparison of 802.11’s

Standard 802.11a 802.11b 802.11g

Number of channels

Interference

Bandwidth

Power consumption

Range/penetration

Upgrade/compatibility

Price

http://www.nwfusion.com/techinsider/2002/0520wlan/0520feat1.html

Page 94: 1 Wireless Networks Ivan Marsic Rutgers University.

94

Route Discovery in DSR (1)

FB

D

E G

H

KJ

I

Y

Z

LA

C

Page 95: 1 Wireless Networks Ivan Marsic Rutgers University.

95

Route Discovery in DSR (2)

Broadcast Tx

Represents a node that has received RREQ for H from C

F

C

B

D

E G

H

KJ

I

Y

Z

LA

RREQ[C]

RREQ[C]

Page 96: 1 Wireless Networks Ivan Marsic Rutgers University.

96

Route Discovery in DSR (3)

F

C

B

D

E G

H

KJ

I

Y

Z

LA

RREQ[C, E]

RREQ[C, B]

Page 97: 1 Wireless Networks Ivan Marsic Rutgers University.

97

Route Discovery in DSR (4)

F

C

B

D

E G

H

KJ

I

Y

Z

LA

RREQ[C, E, G]

RREQ[C, B, A]

RREQ[C, B, A]

Page 98: 1 Wireless Networks Ivan Marsic Rutgers University.

98

Route Discovery in DSR (5)

F

C

B

D

E G

H

KJ

I

Y

Z

LA

RREP[C, E, G, H]

RREQ[C, B, A, K]

Page 99: 1 Wireless Networks Ivan Marsic Rutgers University.

99

Route Discovery in AODV (1)

F

C

B

D

E G

H

KJ

I

LA

Page 100: 1 Wireless Networks Ivan Marsic Rutgers University.

100

Multihop Throughput

Challenge: more hops, less throughputLinks in route share radio spectrumExtra hops reduce throughput

Throughput = 1

Throughput = 1/2

Throughput = 1/3

Page 101: 1 Wireless Networks Ivan Marsic Rutgers University.

101

Cellular Hierarchy

High-tier

Low-tier

Satellite

High Mobility Low MobilityWide Area

Regional Area

Local Area

High-tier

Low-tier

Satellite

High Mobility Low MobilityWide Area

Regional Area

Local Area

Page 102: 1 Wireless Networks Ivan Marsic Rutgers University.

102

Hybrid Wireless Networks

Infrastructure + MANET

(a) (b) (c)

Access Point

Mobile NodeObstacle

Page 103: 1 Wireless Networks Ivan Marsic Rutgers University.

103

Community Mesh Network

WiMAXTower

Mesh NetworkDSLCable NetworkWiMAXFiber Backbone

CableOperator

CentralOffice

Internet