of 58
7/28/2019 Random AccessCSMA
1/58
Topics on Wireless Ad Hoc Networks
Random Access
Dr. Stavros ToumpisTelecommunications Research Center Vienna
1
7/28/2019 Random AccessCSMA
2/58
Medium Access Control (MAC)
Nodes must decide when to access the channel, i.e., transmit.
Two conflicting targets:
Collisions of packets must be avoided. Bandwidth must not be underutilized.
Therefore, a balance is needed. This is the task of the MAC protocol.
2
7/28/2019 Random AccessCSMA
3/58
Classification of MAC Protocols
Random Access: Nodes contend for the channel whenever they have a packet. Simplest example is Slotted Aloha: Nodes transmit packets whenever
they arrive.
Transmission Scheduling:
Each node can transmit during a preassigned set of slots. Simplest example is TDMA/FDMA/CDMA.
Hybrid Protocols:
Nodes have preassigned slots but also contend. Simplest example is Reservation Aloha: Nodes contend for a slot
whenever they need it, but then have priority in following frames.
Disclaimer: Classification not perfect (research area is vast)
3
7/28/2019 Random AccessCSMA
4/58
Advantages of Each Approach
Random access is preferable when:
traffic is bursty (e.g., data). topology changes fast or is unknown.
Transmission scheduling is preferable when:
traffic is periodic (e.g. voice traffic). topology changes slowly. quality guarantees are needed.
Hybrid protocols are adaptable to the traffic/mobility conditions.
4
7/28/2019 Random AccessCSMA
5/58
In this lecture: Random Access Protocols
Before CSMA/CA:
Slotted Aloha (Roberts 72 [1])
Carrier Sense Multiple Access (CSMA) (Kleinrock 72 [2])
CSMA with Collision Detection (CSMA/CD) (Metcalfe 76 [3])
Busy Tone Multiple Access (BTMA) (Tobagi 76 [4])
CSMA with Collision Avoidance (CSMA/CA):
MACA (Karn 90 [5])
MACAW (Bharghavan 94 [6])
IEEE 802.11 standard (97 [7])
After CSMA/CA:
Fairness enhanced CSMA/CA (Ozugur 99 [8])
Dual Busy Tone Multiple Access (DBTMA) (Haas 99 [9, 10])
CSMA/CA with Power Control (Agarwal 01 [11], Jung 02 [12]) CATA (Tang 99 [13])
Progressive Backoff Algorithm (PBOA), Progressive Ramp Up Algorithm (PRUA)
(Toumpis 03 [14])
5
7/28/2019 Random AccessCSMA
6/58
Slotted Aloha [1]
Time is slotted.
Each node transmits whenever a packet arrives.
At end of slot nodes know whether there was a collision.
If there was a collision, nodes backs off for random time.
Nodes do not listen before transmitting, for one or more of the followingreasons:
other transmitters are hidden from them, so even if they listen theywill not hear anything (as in satellite communications). propagation delay is very large, so the information they will gain will
be irrelevant. it is not worth it, because traffic is very small.
6
7/28/2019 Random AccessCSMA
7/58
Slotted Aloha Example
1
2
3
45
6
n
PACKET
ARRIVALS
SLOTS
TIME
TIME
7
7/28/2019 Random AccessCSMA
8/58
Carrier Sense Multiple Access (CSMA) [2]
Nodes monitor the channel (i.e., there is channel sensing).
A node with a packet transmits only if it perceives an idle channel.
If a node with a packet perceives a busy channel, it waits for the channelto become available, then waits for a random time interval (to avoidcollisions).
If there is a collision, packet is backlogged for a random time interval.
Do humans use CSMA?
8
7/28/2019 Random AccessCSMA
9/58
CSMA Example
A,B,C,D are along a straight line, and within range of each other.
A
B
C
CARRIER SENSING
DATA FOR C
D DATA FOR C
Vulnerable window Time
Packet Arrivals
AB D
CSMA works great in totally connected networks with small propagationdelays. (It works perfectly in totally connected networks with zeropropagation delays.)
9
7/28/2019 Random AccessCSMA
10/58
CSMA with Collision Detection (CSMA/CD) [3]
Nodes sense the channel before and after transmitting.
If a collision is detected, all transmitters stop immediately.
Collisions are detected much earlier, and do not cost a lot of bandwidth.
Ethernet is based on CSMA/CD.
Unfortunately, CSMA/CD is not practical in wireless networks, becauseit is very hard to transmit and listen at the same time.
So we can not have wireless Ethernet.
10
7/28/2019 Random AccessCSMA
11/58
CSMA/CD Example
A,B,C,D are along a straight line, and within range of each other.
Backoff
CARRIER
SENSING
CARRIER
SENSING
Backoff
A
B
C
D
DATA FOR C
Vulnerable window
Time
Packet ArrivalsB D A
DATA FOR C
11
7/28/2019 Random AccessCSMA
12/58
Performance of CSMA in Distributed Topologies
It is possible that a node sense the channel to idle, but should nottransmit (the hidden terminal problem).
It is possible that a node senses the channel busy, but should transmit(the exposed terminal problem).
A
D
C
B
A
D
C
B
2) Exposed Terminal Problem1) Hidden Terminal Problem
(In the examples, only nodes connected by a straight line can listen to eachothers transmissions.)
12
7/28/2019 Random AccessCSMA
13/58
7/28/2019 Random AccessCSMA
14/58
BTMA Example
B
D
Vulnerable window
CARRIER SENSING
BT SENSING
BUSY TONE
DATA
Time
A
C
(Only nodes connected by a straight line can receive each others signals.)
14
7/28/2019 Random AccessCSMA
15/58
Disadvantages of BTMA
Some bandwidth is sacrificed, and nodes must be full-duplex (harder than half-duplex).
If all receivers transmit BT, some transmitters are unnecessarily stopped.
If only intended receiver transmits BT, there are collisions elsewhere.
A
D
A C
D
E
E
1) All receivers transmit BT 2) Intended receiver transmits BT
C
BB
BLUE: DATA CHANNEL
RED: CONTROL CHANNEL
In first case, C can not transmit a packet to A (he should have been allowed).
In second case, B transmits a packet to C and there is collision (bandwidth is wasted).
15
7/28/2019 Random AccessCSMA
16/58
CSMA with Collision Avoidance (CSMA/CA) (NOT
Colvin 83 [15])
RTS CTS
2) CTS (Clear To Send)
3) DATA 4) ACK (Acknowledge)
1) RTS (Request To Send)
A D
C
B
A D
C
B
ACK
A D
C
B
DATA
A D
C
B
One way to think about it: BTMA in time domain.
ACK packet exists only in some incarnations (for example IEEE 802.11).
16
7/28/2019 Random AccessCSMA
17/58
Carrier Sense and Virtual Carrier Sense
Nodes sense the channel for the carrier, as in plain CSMA.
Nodes keep track of CTS packets they received: virtual carrier sense.
A
B
C
D
RTS
Tim
Vulnerable window
CTS
CS VCS
CS
DATA
ACK
VCS CS
CS
VCS
17
7/28/2019 Random AccessCSMA
18/58
Backoff Mechanism
Motivation:
When channel becomes available, maybe more than one nodes havepackets to transmit.
If all of them transmit, there will be collisions.
Solution: nodes backoff for random times.
Each selects a random number for backoff counter between 1 and thecontention window CW.
While channel is idle, nodes reduce backoff counter.
First one to hit0
transmits. The rest freeze their counters.
18
7/28/2019 Random AccessCSMA
19/58
Backoff Countdown Example
Time
X
Y
B =30
COUNT DOWN
DATACOUNT DOWN
B =0
B =10
COUNTER
FREEZE
DATA
COUNT
DOWN
FREEZE COUNTERDOWN
COUNT
B =10
B =25 B =15
1 1 1 B =0
B =202 2 22
1
19
7/28/2019 Random AccessCSMA
20/58
Congestion Control
The contention window size should reflect the level of congestion in thenetwork:
If there is little traffic, the contention window should be small. If there is a lot of traffic, the window should be large to reduce
collisions.
The solution of IEEE 802.11:
If a node transmits but fails, it doubles the contention window, untilit reaches CWmax.
If a node transmits and succeeds, it resets the contention window toCWmin.
The solution of MACAW:
Multiplicative Increase, Linear Decrease (MILD). The contention window does no change very fast.
20
7/28/2019 Random AccessCSMA
21/58
PROBLEM 1: The Fairness Problem
AD
C
B
A
B
C
D
E
F
Assume heavy traffic: all nodes need the channel all of the time.
But red and blue transmissions cannot coexist (see previous slide).
In 4-node example, ifA has the channel, D can not get it back.
We have long-term fairness but not short-term fairness.
In 6-node example, C can only initiate transmission if both A and F aresilent at the same time.
Backoff algorithm makes matters worse!
Nodes that fail in handshake double their contention window.
21
7/28/2019 Random AccessCSMA
22/58
Solutions of Fairness Problem
Add extra control packets (Bharghavan et al., 1994 [6]):
Receivers invite transmitters when they sense that the channel becomesavailable.
Works great in 4-node example. Does not work in 6-node example.
Change backoff algorithm (Ozugur et al., 1999, [8]):
Nodes that have more than their fair share of the channel use greaterlarger backoff counter.
But what is a fair share of the channel?
There is no nice solution on the MAC layer.
Routing layer should help.
22
7/28/2019 Random AccessCSMA
23/58
Performance of [8]
Fairness Index is ratio of maximum to minimum throughput.
Fairness is still not achieved, and bandwidth is lost.
23
7/28/2019 Random AccessCSMA
24/58
PROBLEM 2: Spatial Reuse with CSMA/CA
CSMA/CA ensures fewer collisions, but introduces artificial restrictionson transmissions.
In following examples, the blue and the red transmission arecompatible, but CSMA/CA does not allow them.
Reason: With CSMA/CA, transmitter is also required to receive,
receiver is also required to transmit.
Hidden terminal problem morphs to another problem!
Exposed terminal problem not solved at all!
1) Receivers cannot be neighbors
A D
C
B
A D
C
B
2) Transmitters can not be neighbors
24
7/28/2019 Random AccessCSMA
25/58
Dual Busy Tone Multiple Access (DBTMA) [9, 10]
Bandwidth is divided in 4 channels:
Data channel (for DATA packets). Control Channel (for RTS and CTS packets). Channel for transmitting BTr (receiver busy tone).
Channel for transmitting BTt (transmitter busy tone).
C
BLEGEND
BTr
BTt
DATA
RTS/CTS
25
7/28/2019 Random AccessCSMA
26/58
Evaluation of DBTMA
Disadvantages:
Complicated. Bandwidth is sacrificed for control channel and busy tone channels.
Advantages: Data packets never collide (this is possible with CSMA/CA, and very
wasteful). Artificial restrictions on transmissions of CSMA/CA are lifted:
1) Receivers can be neighbors
A D
C
B
A D
C
B
2) Transmitters can be neighbors
26
7/28/2019 Random AccessCSMA
27/58
PROBLEM 3: CSMA/CA has no Power Control
Transmitter receiver pairs must use the same power becausecommunication must be bidirectional (setting (A)).
Different pairs must use same power, otherwise strong transmitters killweak transmitters (setting (B)).
(A) (B)
27
7/28/2019 Random AccessCSMA
28/58
First Try (Agarwal 01 [11])
Have all the nodes transmit the RTS/CTS packets with maximum power, and theDATA/ACK packets with less power.
Transmissions are protected, power is conserved.
Actually a bad idea:
Still nodes must be separated more than necessary.
It is harder for potential interferers to sense the channel.
Following figure shows the performance of this protocol (BASIC) on chain of 50nodes:
28
7/28/2019 Random AccessCSMA
29/58
Second Try (Jung 02 [12])
Same as previous, but now nodes transmit power spikes during the datapackets.
Throughput same as IEEE 802.11: each interferer appears to betransmitting with maximum power ...
... but energy efficiency is better, because spikes are short.
29
7/28/2019 Random AccessCSMA
30/58
PROBLEM 4: Transmissions over weak links are not
protected
Interference Radius
11
CTS Radius
Anyone in outer ring will not hearCTS but can destroy the transmission
Interference Radi
CTS Radius
T
100
R
T
R
(Minimum SINR required for successful reception: T = 10.)
PCM also suffers from this!
Problem is typically masked by poor physical layer mode.
30
7/28/2019 Random AccessCSMA
31/58
PROBLEM 5: First In First Out (FIFO) is suboptimal
C
A B
D
A A AB B B
C
C C C
D
D DD
A B
nonFIFO
First In First Out
D
C
A A
C A
A
D
CC
Dropping the FIFO requirement can lead to a tighter packing of transmissions.
Problem is typically masked by simple traffic models.
31
7/28/2019 Random AccessCSMA
32/58
Other Problems
Control packets might collide.
Nodes must decide a power threshold for declaring the channel busy.
Placing the threshold too high will prohibit some transmissions thatcould take place.
Placing it too low will not protect sufficiently other transmissions. No matter where you place the threshold, mistakes will be made. The problem is that you are sensing at the wrong place (i.e., the
transmitter, not the receiver.)
32
7/28/2019 Random AccessCSMA
33/58
An Improvement: Slotted Time (Tang 99, [13])
Time
RTS CTS DATA
A D
C
B
A D
C
B
STEP 1 (RTS), STEP 3 (DATA) STEP 2 (CTS)
33
7/28/2019 Random AccessCSMA
34/58
Even better: Multiple Minislot Pairs (Toumpis 03 [14])
PERIOD DATA SLOT
RTSm
CTSm
RTS1
RTS2
CTS2
CTS1
FRAME n1 FRAME n FRAME n+1
CONTENTION
Many designs possible: Progressive Back Off Algorithm. Progressive Ramp Up Algorithm.
34
7/28/2019 Random AccessCSMA
35/58
Progressive Back Off Algorithm: Overview
Initially, all nodes with packets contend.
Nodes that are being unsuccessful:
Either back off (so others will have a better chance).
Or remain in contention, but pick a new destination (if such exists).
Nodes that succeed, use the rest of the slots for power control.
Energy is conserved. Interference is reduced for the rest.
35
7/28/2019 Random AccessCSMA
36/58
Progressive Back Off Algorithm: The Rules
Nodes are divided in three groups: Contending, locked, silent.
At the start of the first RTS minislot, nodes with packets form the contending group,
the rest form the silent group.
At the start of the i-th RTS minislot:
Silent nodes listen to the channel.
Contending nodes transmit to potential destination with maximum power.
Locked nodes transmit to destination (with power specified at previous slot).
At the start of the i-th CTS minislot:
Contending and locked nodes remain silent.
Silent nodes that received an RTS packet from a contending node in the previousminislot transmit a CTS packet, specifying new power for the transmitter.
Silent nodes that received an RTS packet from a locked node transmit a CTS
packet only if SINR was greater than (1 + )T.
At the end of the i-th CTS minislot: Contending nodes that received a CTS become locked.
Contending nodes that did not receive a CTS:
With probability p remain contending, but select new destination.
With probability 1 p become silent.
36
7/28/2019 Random AccessCSMA
37/58
An example network
0 50 100 150 200 250 300 350 400
0
50
100
x (m)
y(m)
1
2
3
4
5
6
78
9
1011
12
Reception is successful as long as the SINR is greater than a threshold T = 10 dB.
All transmitters transmit with rate R = 1 Mbps.
Maximum transmitter power is Pmax = 0.3 W.
Power gains decay exponentially with distance, with decay exponent = 4, and thereis no fading:
Gij = Kdij .
Nodes that can communicate directly in the absence of interference are connected
by a line in the figure.
37
7/28/2019 Random AccessCSMA
38/58
Three Routing Protocols: RP-10, RP-30, RP-120
Discard weak links.
SNR in the absence of interference is below 10 (for RP-10), 30 (forRP-30), or 120 (for RP-120).
Keep the rest of the links. Use them to construct minimum hop routes.
Different tradeoffs:
RP-10 needs few hops.
RP-120 is robust to interference. RP-30 is balanced.
38
7/28/2019 Random AccessCSMA
39/58
Routing Tables
RP10
83
121
2 10
6
4
5
9
7
11
RP30
83
121
2 10
6
4
5
9
7
11
RP120
83
121
2 10
6
4
5
9
7
11
39
7/28/2019 Random AccessCSMA
40/58
Operation of PBOA under RP-120 (RTS1-CTS1)
! !
! !
" " "
" " "
# #
# #
$ $
$ $
% %
% %
& &
& &
' '
' '
( (
( (
) )
) )
0 0
0 0
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6
6 6
7 7
7 7
8 8
8 8
9 9
9 9
@ @
@ @
A A
A A
B B
B B
C C
C C
D D
D D
E E
E E
F F
F F
G G
G G
H H
H H
I I
I I
P P
P P
Q Q
Q Q
10
RTS 1
CTS 1
10
Contending: Locked: Silent:
121
2 10113
6
7
8
4
5
9
121
2 1011
93
6
7
8
4
5
1010
10
1010
40
7/28/2019 Random AccessCSMA
41/58
Operation of PBOA under RP-120 (RTS2-CTS2)
! !
! !
" " "
" " "
# #
# #
$ $
$ $
% %
% %
& &
& &
' '
' '
( (
( (
) )
) )
0 0
0 0
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6
6 6
7 7
7 7
8 8
8 8
9 9
9 9
@ @
@ @
A A
A A
B B
B B
C C
C C
10
RTS 2
CTS 2
10
Contending: Locked: Silent:
1010
121
2 10113
6
7
8
4
5
9
121
2 1011
93
6
7
8
4
5
5 10107
41
7/28/2019 Random AccessCSMA
42/58
Operation of PBOA under RP-120 (RTS3-CTS3)
! !
! !
" " "
" " "
# #
# #
$ $
$ $
% %
% %
& &
& &
' '
' '
( (
( (
) )
) )
0 0
0 0
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6
6 6
7 7
7 7
8 8
8 8
9 9
9 9
RTS 3
CTS 3
Contending: Locked: Silent:
1010
121
2 10113
6
7
8
4
5
9
121
2 1011
93
6
7
8
4
5
4 1086
42
7/28/2019 Random AccessCSMA
43/58
Parameter Selection:
Throughput versus persistence probability p
0 0.2 0.4 0.6 0.8 10
0.1
0.2
0.3
0.4
0.5
UniformCap
acity,
Cu(Mbps)
Persistence Probability, p
(a)
(b)
(c)
(a) PBOA with 10 slots.
(b) PBOA with 5 slots.
(c) PBOA with 2 slots.
With more slots, nodes should be more persistent.
Intuition: Backing off should be progressive over all slots.
43
7/28/2019 Random AccessCSMA
44/58
Progressive Ramp Up Algorithm: Overview
PRUA works in the opposite way from PBOA.
In the beginning of the contention period, nobody transmits.
As the contention period progresses, every now and then a node will tryto grab the channel.
Successful nodes persist, unsuccessful ones may try again later.
Nodes that do not transmit, monitor the channel to gain informationabout the competition and make educated decisions.
Nodes pick destinations for which the conditions appear to be mostfavorable.
A transmission schedule is slowly being built.
44
7/28/2019 Random AccessCSMA
45/58
Progressive Ramp Up Algorithm: The Rules
At the start of the i-th RTS minislot a node A will transmit an RTS packet:
If it transmitted an RTS in the previous RTS minislot and heard a CTS packet in
reply.
Or all of the following conditions are satisfied: A did not transmit a CTS packet in the previous minislot pair.
The received power in the previous CTS minislot did not exceed a threshold PT.
If A has not decoded an RTS in the previous RTS minislot, it must have anon-empty queue.
If A has decoded an RTS transmitted from a node B in the previous RTS minislot,
it must have a packet for a node C such that C can decode the packet in the
presence of interference from node B.
A must perform a biased coin toss, with probability p, and succeed.
At the start of the i-th CTS minislot, whoever received an RTS packet addressed tohim, replies with a CTS packet.
At the start of the data slot, whoever received a CTS packet at the last CTS minislot
transmits a data packet.
45
7/28/2019 Random AccessCSMA
46/58
Operation of PRUA under RP-120 (RTS1-CTS1)
! !
! !
" "
" "
# #
# #
$ $ $
$ $ $
% %
% %
& &
& &
' '
' '
( ( (
( ( (
) )
) )
0 0
0 0
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6
6 6
7 7
7 7
8 8
8 8
9 9
9 9
@ @
@ @
A A
A A
B B
B B
C C
C C
D D
D D
E E
E E
F F
F F
G G
G G
H H
H H
I I
I I
P P
P P
Q Q
Q Q
RTS 1
CTS 110
Nodes with no packets: Nodes with packets:
121
2 10113
6
7
8
4
5
121
2 1011
93
6
7
8
4
510
9
46
7/28/2019 Random AccessCSMA
47/58
Operation of PRUA under RP-120 (RTS2-CTS2)
! !
! !
" "
" "
# #
# #
$ $ $
$ $ $
% %
% %
& &
& &
' '
' '
( ( (
( ( (
) )
) )
0 0
0 0
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6
6 6
7 7
7 7
8 8
8 8
9 9
9 9
@ @
@ @
A A
A A
B B
B B
C C
C C
D D
D D
E E
E E
F F
F F
G G
G G
H H
H H
I I
I I
P P
P P
Q Q
Q Q
RTS 2
CTS 210
Nodes with no packets: Nodes with packets:
121
2 10113
6
7
8
4
5
121
2 1011
93
6
7
8
4
510
9
10
10
47
7/28/2019 Random AccessCSMA
48/58
Operation of PRUA under RP-120 (RTS3-CTS3)
! !
! !
" "
" "
# #
# #
$ $ $
$ $ $
% %
% %
& &
& &
' '
' '
( ( (
( ( (
) )
) )
0 0
0 0
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6
6 6
7 7
7 7
8 8
8 8
9 9
9 9
@ @
@ @
A A
A A
B B
B B
C C
C C
D D
D D
E E
E E
F F
F F
G G
G G
H H
H H
I I
I I
P P
P P
Q Q
Q Q
RTS 3
CTS 3
Nodes with no packets: Nodes with packets:
10 121
2 10113
6
7
8
4
5
121
2 1011
93
6
7
8
4
510
9
10
10
10
10
48
7/28/2019 Random AccessCSMA
49/58
The Performance of PBOA, PRUA
0 0.1 0.2 0.3 0.4 0.5 0.60
0.1
0.2
0.3
0.4
0.5
0.6
r3,12
r4,
8
(a)
(b)
(c)
(d)
(e)(f)
(g)
RP10 RP30 RP1200
0.2
0.4
0.6
0.8
1
Routing Protocol
Uniformc
apacity,
Cu(
Mbps) (b)
(a)
(c)
(d)
(e)
(f)
(g)
(a) Perfect power control. (e) IEEE 802.11.
(b) ON/OFF power control. (f) PBOA.(c) Optimal PC. (g) PRUA.(d) ON/OFF PC.
(Lines (c)-(g) of left figure are with RP-120.)
49
7/28/2019 Random AccessCSMA
50/58
Energy Efficiency
0 0.1 0.2 0.3 0.4 0.5 0.60
2
4
6
8
10
12
14
16
18
Throughput, T (Mbps)
EnergyperPack
et,Ep(mJoule)
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(a) IEEE 802.11, RP-10.
(b) PBOA, RP-10.
(c) PRUA, RP-10.
(d) IEEE 802.11, RP-30.
(e) PBOA, RP-30.(f) PRUA, RP-30.
(g) IEEE 802.11, RP-120.
(h) PBOA, RP-120.
(i) PRUA, RP-120.
50
7/28/2019 Random AccessCSMA
51/58
Throughput-Delay Curves (with RP-120)
0 0.1 0.2 0.3 0.4 0.5 0.60
100
200
300
400
500
600
700
800
900
1000
Delay,D
(msec)
Throughput, T (Mbps)
(a) (b)
(c)
(d)
(e)
(f)
(a) IEEE 802.11.
(b) PBOA.
(c) PRUA.
(d) Uniform capacity, ON/OFF power
control.(e) Uniform capacity, Optimal power
control.
(f) Bound on delay due to packetizing of
data.
51
7/28/2019 Random AccessCSMA
52/58
Parameter Selection:
Throughput versus number of minislot pairs m
5 10 15 20 250.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
UniformCapacity,
Cu
(Mbps)
Minislot pairs, m
(a)
(b)
(c)
(d)
(e)
(f)
(a) PBOA, RP-10.
(b) PBOA, RP-30.
(c) PBOA, RP-120.
(d) PRUA, RP-10.
(e) PRUA, RP-30.
(f) PRUA, RP-120.
52
7/28/2019 Random AccessCSMA
53/58
Performance of PBOA and PRUA
3.5 4 4.5 5 5.50
0.5
1
1.5
2
2.5
3
3.5
4
Throughput, T (Mbps)
EnergyperPacket,E
p(mJoule)
IdealPCM
CSMA/CA
PBOA
PRUA
Ideal-PCM is CSMA/CA with the following modification:
Nodes only transmit with the minimum power required for successful reception.
Interferers seem to transmit with maximum power.
53
7/28/2019 Random AccessCSMA
54/58
Why we still can not achieve capacity?
An example network
Network Model:
Nodes connected by lines can communicate directly, the rest do not interfere.
Receivers can decode at most one packet at a time (any two partially overlapping
packets will both be destroyed).
Transmitter rate: R = 1 Mbps.
Maximum number of simultaneous transmissions s: In theory: s = 4.
With CSMA/CA: s = 2.
With PBOA or PRUA: s = 4.
54
7/28/2019 Random AccessCSMA
55/58
Uniform Capacity of Example Network
1
2
3
4
Uniform capacity calculation: Simultaneous transmissions: s = 4.
Average number of hops: h = 167 .
Uniform capacity is Cu = Rsh =
74 Mbps = 1.75 Mbps.
Uniform capacity of MAC protocols (by simulation):
CCSMA/CAu = 0.635 Mbps.
CPBOAu = 1.05 Mbps. CPRUAu = 1.145 Mbps.
To achieve capacity, arbitrarily distant nodes must coordinate. Impossiblewith distributed MAC protocols!
55
7/28/2019 Random AccessCSMA
56/58
Conclusions
We studied random access protocols:
Before CSMA/CA (i.e. RTS/CTS handshake) CSMA/CA (very important due to its use in IEEE 802.11) After CSMA/CA
Emphasis on why things do not work, how we try to make them work,and how we always fail in some way or another.
Capacity is lost in many different places:
Power control. Routing protocol. Queuing discipline. (Distributed) Medium Access Control.
56
7/28/2019 Random AccessCSMA
57/58
References
[1] L. G. Roberts, Aloha packet system with and without slots and capture, Stanford, CA: StanfordResearch Institute, Advanced Research Projects Agency, Network Information Center.
[2] L. Kleinrock and F. A. Tobagi, Packet switching in radio channels: Part ICarrier Sense MultipleAccess models and their throughput-delay characteristics, IEEE Trans. Commun., vol. 23, no. 12, pp.14001416, Dec. 1975.
[3] R. M. Metcalfe and D. R. Boggs, Ethernet: distributed packet switching for local computer networks,Comm. of the ACM, vol. 19, no. 5, pp. 395404, July 1976.
[4] F. A. Tobagi and L. Kleinrock, Packet switching in radio channels: Part IIthe hidden terminalproblem in carrier sense multiple-access and the busy tone solution. IEEE Trans. Commun., vol. 23,
no. 12, pp. 14171433, Dec. 1975.
[5] P. Karn, MACA: A new channel access method for packet radio, in Proc. CNC, vol. 1, London,Ontario, Canada, Sep. 1990, pp. 134140.
[6] V. Bharghavan, A. Demers, S. Shenkar, and L. Zhang, MACAW: A media access protocol for wirelessLANs, in Proc. ACM SIGCOMM, London, UK, Aug. 1994, pp. 212225.
[7] Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specification, IEEE Std.802.11.
[8] T. Ozugur, M. Naghshineh, P. Kermani, and J. A. Copeland, Fair media access for wireless LANs, inProc. IEEE GLOBECOM, vol. 1B, Rio De Janeiro, Brazil, Dec. 1999, pp. 570579.
[9] Z. J. Haas, J. Deng, and S. Tabrizi, Collision-free medium access control scheme for ad-hoc networks,in Proc. IEEE MILCOM, vol. 1, Atlantic City, NJ, Nov. 1999, pp. 276280.
57
7/28/2019 Random AccessCSMA
58/58
[10] Z. J. Haas and J. Deng, Dual busy tone multiple access (DBTMA) a multiple access control schemefor ad hoc networks, IEEE Trans. Commun., vol. 50, no. 6, pp. 975985, June 2002.
[11] S. Agarwal, S. Krishnamurthy, R. H. Katz, and S. K. Dao, Distributed power control in ad hoc wirelessnetworks, in Proc. IEEE PIMRC, 2001.
[12] E.-S. Jung and N. H. Vaidya, A power control MAC protocol for ad hoc networks, in Proc. ACMMobicom, Atlanta, GA, Sep. 2002, pp. 3647.
[13] Z. Tang and J. J. Garcia-Luna-Aceves, A protocol for topology-dependent transmission scheduling inwireless networks, in Proc. IEEE WCNC, vol. 3, New Orleans, LA, Sep. 1999, pp. 13331337.
[14] S. Toumpis and A. J. Goldsmith, Performance, optimization, and cross-layer design of media accesscontrol protocols for wireless ad hoc networks, in Proc. IEEE ICC, Anchorage, AK, May 2003, pp.22342240.
[15] A. Colvin, CSMA with collision avoidance, Computer Communications, vol. 6, no. 5, pp. 227235,May 1983.
[16] B. P. Crow, I. Widjaja, J. G. Kim, and P. T. Sakai, IEEE 802.11 Wireless Local Area Networks, IEEECommun. Mag., vol. 35, no. 9, pp. 116126, Sep. 1997.
[17] D. P. Bertsekas and R. Gallager, Data Networks, 2nd ed. Englewood Cliffs, NJ: Prentice Hall, Dec.1991.