Fair Sharing of MAC under TCP in Wireless Ad Hoc Networks Mario Gerla Computer Science Department...

Fair Sharing of MAC under TCP in Wireless Ad Hoc Networks

Mario Gerla

Computer Science Department

University of California, Los Angeles

Los Angeles, CA 90095

http://www.cs.ucla.edu/NRL/wireless

UCLA DARPA Domains Project

Outline

• Overview of CSMA, FAMA and IEEE 802.11.

• MAC performance with TCP. Variable Hop Length Experiments. Hidden Terminal Experiments. Ring Experiments. Grid Experiments.

• Static.• Mobility.


Simulation Using GloMoSim

• Detailed model of the protocol stack.• Allows investigation of TCP and MAC layer

interactions.• Capability to simulate large number of nodes.• GloMoSim web page.

http://pcl.cs.ucla.edu/projects/domains/glomosim.html


MAC Layer Protocols

• CSMA Requires carrier sensing before transmission. If the channel is free, the packet is transmitted immediately. Otherwise, it is rescheduled after a random timeout.

• FAMA Builds on CSMA. Uses the RTS (Request To Send) and CTS (Clear To Send) exchange

to prepare the floor for data transmission.

• 802.11 Uses carrier sensing and RTS/CTS, similar to FAMA. Utilizes link-level ACKs. Collision Avoidance scheme.


Variable Hop Length Experiments Configuration

0 1 2 543

• Each node is 10 meters apart from its neighbors.• Each node has a radio power range of 10 meters.• 2Mbps channel bandwidth.

• FTP traffic.• TCP window size varies from 1 to 16 packet size.• Variable number of hops (single connection).

i.e., FTP connection 0-1, 0-2, 0-3, 0-4, 0-5 (one at a time).


Variable Hop Length Experiments ResultsVariable Hop Length Experiments Results

Variable Number of Hops Experiment (CSMA)

0500000

100000015000002000000

1 2 3 4 5

Number of Hops

Thro

ughp

ut (b

ps)

W = 1

W = 2

W = 4

W = 8

W = 16


Variable Hop Length Experiments Results (Cont’d)


Variable Number of Hops Experiment (FAMA)

0500000

100000015000002000000

1 2 3 4 5

Number of Hops

Thro

ughp

ut (b

ps)

W = 1

W = 2

W = 4

W = 8

W = 16




• CSMA and FAMA degrades with window size > 1 pkt. Collisions between TCP data and ACKs.

• 802.11 performs the same no matter the window size. Link-level ACKs combat collisions.

Variable Number of Hops Experiment (802.11)

0500000

100000015000002000000

1 2 3 4 5

Number of Hops

Thro

ughp

ut (b

ps)

W = 1

W = 2

W = 4

W = 8

W = 16


Hidden Terminal Experiments Configuration

Hidden Terminal Experiments Configuration

• FTP traffic

• Connections from node 0 to node 1 and from node 2 to node 1.

• Node 0 and node 2 cannot hear each other.

0 1 2


Hidden Terminal Experiments Results

Hidden Terminal Experiments Results

• CSMA suffers from hidden terminal.

• FAMA and 802.11 performs well due to RTS/CTS exchange.

0

500000

1000000

1500000

2000000

Throughput (bps)

CSMA FAMA 802.11

MAC Protocol

Hidden Terminal Experiment

0-1

2-1


Ring Experiment Configuration

Ring Experiment Configuration

• FTP traffic

• Multiple single hop connections. i.e., 0-1, 1-2, 2-3, 3-4, 4-5, 5-6, 6-7, 7-0 (at the same time).

0

1

6

5

4

3

2

7


Ring Experiment ResultsRing Experiment Results

• FAMA best in both fairness and aggregate throughput.

• 802.11 unfairness due to timers.

0

100

200

300

400

500

600

Th

rou

gh

pu

t (K

bp

s)

CSMA FAMA 802.11

MAC Protocol

0-1

1-2

2-3

3-4

4-5

5-6

6-7

7-0


802.11 Fairness802.11 Fairness

• Yield time. Time a node yields after transmitting a frame.

0100200300400500600

Th

rou

gh

pu

t (K

bp

s)

802.11 (Standard) 802.11 (YieldTime)

MAC Protocol

0-1

1-2

2-3

3-4

4-5

5-6

6-7

7-0


Grid Experiment Configuration

• Each node is 10 meters apart from its horizontal and vertical neighbors.• Each node has a radio power range of 30 meters.• FTP connections are established between node 18 to node 26, node 36 to node 44, node 54 to node 62, node 2 to node 74, node 4 to node

76 and node 6 to node 78.

0 1

109

8 2

19 18

17 11

20

74 7372

26

80

3 4

1312

5

22 21

14

23

777675

6

15

7

24

16

25

7978

27 28

3736

35 29

4438

30 31

4039

32

41

33

42

34

43

46 45 47 534948 50 51 52

54 55

6463

62 56

7165

57 58

6766

59

68

60

69

61

70


Grid Experiment Configuration (Cont’d)

Grid Experiment Configuration (Cont’d)

• 2Mbps channel bandwidth.

• Nodes move at a rate of 10 meters per second in a random direction with a probability of 0.5.

• When mobility is not considered, static routing is used.

• When mobility is introduced, Bellman-Ford routing is utilized with routing table updates occurring once every second.


Grid Experiments Results (No Mobility)

• Without mobility CSMA performs poorly due to interference by neighboring streams and by intersecting streams. FAMA fair due to RTS/CTS and less aggressive yield time. 802.11 exhibits capture.

050000

100000150000200000250000

Th

rou

gh

pu

t (b

ps)

CSMA FAMA 802.11

MAC Protocol

No Mobility

18-26

36-44

54-62

2-74

4-76

6-78


Grid Experiments Results (With Mobility)

CSMA and FAMA collapse with mobility due to lack of fast loss recovery facilities. 802.11 still operational.

• Link level ACKs help recover from loss caused by transient nodes.• Capture exists.

0

50000

100000

150000

200000T

hro

ug

hp

ut

(bp

s)

CSMA FAMA 802.11

MAC Protocol

With Mobility

18-26

36-44

54-62

2-74

4-76

6-78


Bluetooth Experiments Gerla, M et al, Tyrrenia Conf,

sept 2000

Bluetooth Experiments Gerla, M et al, Tyrrenia Conf,

sept 2000• Experiment #!: TCP throughput in a single piconet.

Throughput versus the no. of TCP connections. Each TCP connection starts from a different slave on the common piconet, and goes through the access point (BT master).

• Experiment #2: TCP throughput when multiple piconets are used in parallel. Each piconet here supports a separate TCP connection.

• Experiment #3: TCP and IP Telephony in a multiple piconet configuration. IP Telephony uses ACL channel. Question: can TCP and Telephony coexist?


S

LAN

IP backbone

M1

IP router

M2 M3

LAN

IP backbone

IP router

M1 M2 M3

S 1

S 2

S 3

M3

IP backbone

M2

M1

LAN

IP router

S

(a) (b) (c)Fig. 4.


Exp # 1:TCP throughput in Bluetooth (single piconet)Exp # 1:TCP throughput in Bluetooth (single piconet)

Bluetooth - Single Piconet

0.687

0.688

0.689

0.69

0.691

0.692

0.693

0.694

1 2 3 4 5 6 7

No. Of TCP Connections

To

tal

TC

P T

hro

ug

hp

ut

(Mb

ps)

Bluetooth


Exp #2:TCP Throughput in WaveLAN vs BT (multiple

piconets)

Exp #2:TCP Throughput in WaveLAN vs BT (multiple

piconets)Bluetooth vs WaveLan

0

1

2

3

4

5

6

7

1 2 3 4 5 6 7 8 9 10

No. of TCP connections

To

tal

TC

P T

hro

ug

hp

ut

(Mb

ps) Bluetooth

WaveLan


Exp #2 (cont): Throughput of TCP flows

Exp #2 (cont): Throughput of TCP flows

Throughput of Individual TCP flows

0

0.2

0.4

0.6

0.8

1

1.2

1 2 3 4 5 6 7 8 9 10

Flow Number --->

Th

rou

gh

pu

t (M

bp

s) -

-->

WaveLan

Bluetooth


TCP and IP TelephonyTCP and IP Telephony

• Voice carried on the ACL channel (ie, VoIP)

• Four piconets

• In each piconet: 1 TCP and 6 Voice connections

• TCP connections “always on” (file transfers)

• Voice: ON-OFF model; 8Kbps coding rate

• Voice packets: 20ms packetization -> 20 bytes

• With header overhead: voice pkt = 30 bytes


Exp #3: Bluetooth; TCP + VoIP


IP Telephony Delay Distribution

0

0.02

0.04

0.06

0.08

0.1

Delay (ms)--->

Fra

ctio

n o

f P

kts-

-->



IP Telephony Complementary Cumulative Distribution

00.10.20.30.40.50.60.70.80.9

1

2 6 10 14 18 22 24 28 32 36 40 60

Delay (ms) --->

Fra

ctio

n o

f P

kts

--->


Exp #3: WaveLan : TCP + VoIP


IP Telephony Delay Distribution

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

Delay (ms) --->

Fra

cti

on

of

Pk

ts -

-->




IP Telephony Complementary Cumulative Delay Distribution

00.10.20.30.40.5

0.60.70.80.9

1

Delay (ms) --->

Frac

tion

of P

kts

--->

With 750 ms playout buffer, still 5% packets lost!


Simulation: what have we learned?

Simulation: what have we learned?

• Fair sharing in BT across TCP connections (IEEE 802.11 is unfair, “capture”- prone)

• BT aggregate throughput exceeds IEEE 802.11

• BT supports voice well even in heavy TCP load (IEEE 802.11 cannot deliver voice with TCP load)

Fair Sharing of MAC under TCP in Wireless Ad Hoc Networks Mario Gerla Computer Science Department...

Documents

Transcript of Fair Sharing of MAC under TCP in Wireless Ad Hoc Networks Mario Gerla Computer Science Department...