Chap 4 Multiaccess Communication Part 1

1
Chap 4 Multiaccess Communication(Part 1)

• Ling-Jyh Chen

2
Overview

• Ethernet and Wi-Fi are both multi-access
  technologies
• Broadcast medium, shared by many hosts
• Simultaneous transmissions will result in
  collisions
• Media Access Control (MAC) protocol required
• Rules on how to share medium

3
Media Access Control Protocols

• Channel partitioning
• Divide channel into smaller pieces (e.g., time
  slots, frequency)
• Allocate a piece to node for exclusive use
• E.g. Time-Division-Multi-Access (TDMA) cellular
  network
• Taking-turns
• Tightly coordinate shared access to avoid
  collisions
• E.g. Token ring network
• Contention
• Allow collisions
• recover from collisions
• E.g. Ethernet, Wi-Fi

4
Contention Media Access Control Goals

• Share medium
• If two users send at the same time, collision
  results in no packet being received
  (interference)
• If no users send, channel goes idle
• Thus, want to have only one user send at a time
• Want high network utilization
• TDMA doesnt give high utilization
• Want simple distributed algorithm
• no fancy token-passing schemes that avoid
  collisions

5
Evolution of Contention Protocols
Aloha

• Developed in the 1970s for a packet radio network

SlottedAloha
Improvement Start transmission only at fixed
times (slots)
CSMA Carrier Sense Multiple Access Improvement
Start transmission only if no transmission is
ongoing
CSMA
CD Collision Detection Improvement Stop
ongoing transmission if a collision is detected
(e.g. Ethernet)
CSMA/CD
6
4.2 Idealized slotted multiaccess model

• m transmitting nodes and one receiver
• Slotted system
• packets are of the same length
• each packet requires one time unit for
  transmission
• the reception of each packet starts at an integer
  time and ends before the next integer time

7

• Poisson Arrivalsoverall arrival rate of the
  system ?individual rate of each node ?/m
• Collision or Perfect Reception
• If just one node sends a packet in a given slot,
  the packet is correctly received.
• If two or more nodes send a packet in a given
  time slot, then there is a collision and the
  receiver obtains no information about the
  contents or the source of the transmitted packets.

8

• 0,1,e Immediate Feedback
• Assuming each node obtains feedback from the
  receiver at the end of each slot
• Retransmission of Collisions
• Assuming each packet involved in a collision must
  be retransmitted in some later slot.
• A node with a packet that must be retransmitted
  is said to be backlogged.

9

• Two addition assumptions
• No buffering
• If one packet at a node is currently waiting for
  transmission or colliding with another packet
  during transmission, new arrivals at that node
  are discarded and never transmitted.
• This assumption provides the lower bound to the
  delay for systems with buffering and flow
  control!
• Infinite set of nodes (m8)
• This assumption provides the upper bound!

10
Slotted ALOHA

• The basic idea
• Each unbacklogged node simply transmit a newly
  arriving packet in the first slot after packet
  arrival.
• Slotted ALOHA risks occasional collisions but
  achieves very small delay if collisions are rare.
• Contrast to TDM systems, which avoids collisions
  at the expense of large delays.

11
Collisions in S-ALOHA
12
Slotted ALOHA (cont.)

• When a collision occurs, each node sending one of
  the colliding packets discovers the collision at
  the end of the slot and becomes backlogged.
• Such nodes wait for some random number of slots
  before retransmitting.

13
Slotted ALOHA (cont.)

• Using infinite-node assumption, the total number
  of retx and tx in a given slot is a Poisson
  random variable with parameter G, where Ggt ?.
• The prob. of a successful transmission in a slot
  is
• In equilibrium, the arrival rate, ?, should be
  the same as the departure rate, Ge-G.

14
Slotted ALOHA (cont.)

• Using GNUPlotset xr 05plot xexp(-x)

15
Slotted ALOHA (cont.)

• The MAX departure rate occurs at G1 and is 1/e
  0.368.
• If Glt1, too many idle slots are generated.
• If Ggt1, too many collisions are generated.

16
Slotted ALOHA (cont.)

• Markov Chain for Slotted ALOHA
• State the number of backlogged packets
• Increases by the number of new arrivals
  transmitted by unbacklogged nodes
• Decreases by one each time if a packet is
  transmitted successfully.

17
Slotted ALOHA (cont.)

• qr the prob. of a backlogged node retx in the
  next slot
• i.e., the number of slots from a collision until
  a given node involved in the collision retx is a
  geometric R.V. having value igt1 with prob.
  qr(1-qr)i-1
• qa the prob. of an unbacklogged node transmits a
  packet in the given slot
• i.e. qa1-e-?/m

18
Slotted ALOHA (cont.)

• Qa(i, n) the prob. that i unbacklogged nodes
  transmit packets in a given slot
• Qr(i, n) the prob. that i backlogged nodes
  transmit.

19
Slotted ALOHA (cont.)
20
Slotted ALOHA (cont.)

• Dn drift in state n, i.e. the expected change
  in backlog over one slot time
• G(n) the expected number of attempted
  transmissions in a slot
• If qa and qr are small,

21
Slotted ALOHA (cont.)

• The drift is the difference between the
  throughput curve (Ge-G) and the straight line

22
Slotted ALOHA (cont.)

• Using infinite-node assumption
• Using no-buffering assumption
• 4.2.3 (optional)

23
Unslotted ALOHA

• Unslotted ALOHA (a.k.a. Pure ALOHA) was the
  precursor to slotted ALOHA.
• In Pure ALOHA, each node transmits a new packet
  immediately upon receiving, rather than waiting
  for a slot boundary.
• If a packet is involved in a collision, it is
  retransmitted after a random delay.

24
Collisions in (Pure) ALOHA
25
Unslotted ALOHA (cont.)

• A frame (red frame) will be in a collision if and
  only if another transmission begins in the
  vulnerable period of the frame
• Vulnerable period has the length of 2 frame times

26
Unslotted ALOHA (cont.)

• Since arrivals are independent, Psucce-2G
• Since attempted transmissions occur at rate G(n),
  the throughput Ge-2G
• The MAX throughput of a Pure ALOHA system
  1/(2e), achieved when G0.5.
• If ? is very small and the mean retx time is very
  large, the system can be expected to run for long
  periods w/o major backlog buildup.
• The main adv. of pure ALOHA is that it can be
  used with variable-length packets.

27
Comparison of ALOHA and S-ALOHA