Chapter 13

Multiple
Access

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.1 Multiple-access protocols

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 13.1 Random Access
» Each station has the right to medium without being controlled by any
other station.
» If more than 1 station tries to send, there is an access conflict (collision)
and the frames will be either destroyed or modified.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 13.1 Random Access

MA

CSMA

CSMA/CD

CSMA/CA

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.2 Evolution of random-access methods

» ALOHA, the earliest random access method, developed at university of

Hawaii in earlier 1970s.
» It was designed to be used on a radio (wireless) LAN.
» Data rate= 9600bps

» The medium is sensed before the transmission.

» The method was called Carrier Sense Multiple Access (CSMA).
» The method later evolved into later two parallel methods:
CSMA/CD and CSMA/CA (CD: collision detection, CA: collision Avoidance)

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.3 ALOHA network

» Base Station is the central controller.

» Every station that needs to send a frame to another station first sends it to
base station.
» The base station receives the frame and relays it to the intended destination.
» Base station acts as HOP.
» The uploading transmission (from a station to base) uses modulation with
carrier frequency of 407 MHz.
» The downloading transmission (from base to station) uses carrier
frequency of 413 MHz.
» The potential collisions in this arrangements are obvious.
» Two more stations may send data at the same time.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.4 Procedure for ALOHA protocol

» ALOHA protocol is simple and based on the following rules:

» MULTIPLE ACCESS: Any station sends a frame when it has a frame to send.
» ACKNOWLEDGEMENT: After sending the frame, the station waits for the
acknowledgement (explicit/implicit). If it does not receive the ACK during the
allotted time, which is 2 times the the maximum propagation delay, it
assumes that the frame is lost.
» After the frame is lost, the station re-sends the frame after a random amount
of time.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.4 Procedure for ALOHA protocol

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 CSMA

» Carrier Sense Multiple Access.

» To minimize the chance of collision and, therefore increase performance
CSMA method was developed.
» The station senses the medium before trying to use it.
» Sense before transmit, or listen before talk.

» CSMA reduces the probability of collision, it CAN NOT eliminate it.

» The possibility of collision exists because of the propagation delay.
» When a station sends a frame it takes a while, although a very short time,
for the first bit to reach every station and for every station to sense it.
» A station may sense the medium and finds it idle, only by propagation by
another station has not yet reached the station.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.5 How Collision happens in CSMA

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 How Collision happens in CSMA

» At time t1, station A senses the medium.

» The medium is idle so it sends a frame.
» At time t2 (t2>t1), station Z senses the medium and also finds it idle, because
at this time propagation from station A has not reached station Z.
» The station Z also sends a frame.
» Two signals collide at time t3 (t3>t2 > t1).
» The result of the collision is the garbled signal, which also propagate in
both directions.
» It reaches station Z at time t4 (t4>t3 > t2 > t1).
» And station A at time t5 (t5>t4 > t3 > t2 > t1).

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.6 Persistence strategies

» Persistence strategies defines the procedures for a station that senses a

busy medium.
» Two strategies have been devised: non persistent and persistent.

» NON-PERSISTENT: A station that has a frame to send senses the line.

» If the line is idle, the station sends immediately.
» If the line is not idle, the station waits for a random period of time and then
senses the line again.
» It reduces chances of collision as it is unlikely that two or more stations wait
for same amount of time and retry again simultaneously.
» However this method reduces the efficiency of the network due to idleness.

» PERSISTENT: A station senses a line, if the line is idle, the station sends a
frame.
» This method has two variations: 1-Persistent and p-persistent.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.6 Persistence strategies

» 1-Persistent: The station senses the medium and if finds it idle, it sends the
frame immediately (with a probability of 1).
» This increases the chances of collision.

» p-persistent: If the station finds the medium idle, the station may or may not
sends the frame.
» It sends with the probability p and refrains from sending with probability 1-p.
» e.g, If p is 0.2, means each station, after sensing the idle line, sends with a
probability of 0.2 (20% of time).
» The station generates a random number between 1-100.
» If the random number is less than 20%, the station will send, otherwise the
station refrains from sending.
» This methods reduces the chances of collision and improves the efficiency.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.6 Persistence strategies

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 CSMA/CD

» Carrier Sense Multiple Access/ Collision Detection.

» Any station can sends a frame, the station then monitors the medium to see
if transmission was successful.
» If so station is finished.
» If fails the station resends, but this time the station waits- it needs to
back off.
» In the exponential back off method, the station waits an amount of time
between 0 and 2N x max propagation time. (the time needed for a bit to reach
the end of the network).
» N value ranges from 1 to onwards in subsequent tries.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 13.7 CSMA/CD procedure

» Carrier Sense Multiple Access/ Collision Avoidance.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 CSMA/CA

» Carrier Sense Multiple Access/ Collision Avoidance.

» There is no collision, this procedure avoids collisions.
» The method uses on of the persistence strategies.
» After it finds the lien idle, the station waits an IFG (Inter Frame Gap) amount
of time.
» It then waits another random amount of time.
» After that, it sends the frame and sets a timer.
» The station waits for an acknowledgement from the receiver.
» If it receives acknowledgement before the timer expires the transmission is
successful, if the receiver doesn’t receives an acknowledgement, the frame
is understood to be damaged or lost.
» The station increments the value of back off parameter, waits for a back off
amount of time, and re-senses the line.
» The CSMA/CA is used in wireless LANs.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.8 CSMA/CA procedure

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 13.2 Control Access
» In controlled access, the stations consult one another to find which
station has the right to send.
» A station cannot send unless it has been authorized by other
stations.
» Three popular techniques are used to control access:

Reservation
Polling
Token Passing

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Reservation access method

» In the reservation access method, a station needs to make a reservation

before sending data.
» Time is divided into intervals.
» In each interval, a reservation frame precedes the data frames sent in that
interval.

» If there are N stations in the system, there are exactly N reservation

mini-slots in the reservation frame.
» When a station needs to send a data frame, it makes a reservation in its own
mini-slot.
» The stations that have made reservation can send their data frames after
the reservation frame.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.9 Reservation access method

» A system with 5 stations and 5 mini-slots reservation frame.

» In the first interval, only stations 1 ,3 and 4 have made reservations.
» In the second interval only station 1 has ,and made a reservation.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 polling

» Polling works with topologies in which one device is designated as

primary station and other devices are secondary stations.
» All data exchange must be made through primary station when the ultimate
destination is a secondary device.
» The primary devices controls the link; the secondary devices follow the
instructions.
» It is up to the primary device to determine which device is allowed to use the
channel.
» The primary device is therefore always the initiator of the session.
» If the primary wants to receive data it asks secondaries to if they have
anything to send; this is called polling.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Select

» Select mode is used whenever a primary device has something to send.

» As the primary controls the links, so when primary is neither sending or
receiving the data, the link is available.
» But what it does not knows is the receiver device is ready to receive.
» So the primary must alert the secondary for the upcoming transmission and
waits to the acknowledgement of the secondary’s ready status.
» Before sending data, the primary creates and transmits a select (SEL) frame.
» One field of which includes address of the intended secondary.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.10 Select

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.11 Poll

» The polling function is used by the primary device to solicit the transmission
from the secondary devices.
» When the primary is ready to receive data, it must ask (poll) each device in
turn if it has anything to send.
» When first secondary is approached it either respond with NAK frame
or with data frame.
» If NAK frame is responded then next secondary is polled by primary.
» When response is data frame then primary reads the frame and ACK it.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.12 Token-passing network

» In token passing method, a station is authorized to send data when it

receives a special frame called a token.
» The stations are arranged in a ring.
» Each station has a predecessor and a successor.
» Frames are coming from predecessor and going to successor.

» When no data are being sent, a token circulates around the ring.
» If a station needs to send data, it waits for the token.
» The station captures the token and send one or more frames.
» When finished it releases the token to successor station.
» Priority and reservation is also used to enhance the procedure.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.13 Token-passing procedure

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 13.3 Channelization

» Channelization is a multiple access method.

» Available bandwidth of the link is shared in time, frequency or
through code, between different stations.
» Three important channelization protocols are as under:

FDMA
TDMA
CDMA

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 FDMA

» Frequency-Division Multiple Access.

» Available bandwidth is shared by all stations.
» Each station has its own allocated band to sends its data.
» The band belongs to the station all the time.
» FDMA is a data-link layer protocol that uses FDM at the physical
layer.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Note:

In FDMA, the bandwidth is divided

into channels.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 TDMA

» Time-Division Multiple Access.

» Entire bandwidth is just one channel.
» The stations shares the capacity of the channel in time.
» Each station has its own allocated time slot to sends its data.
» TDMA is a data-link layer protocol that uses FDM at the physical
layer.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Note:

In TDMA, the bandwidth is just one

channel that is timeshared.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 CDMA

» Code-Division Multiple Access.

» CDMA differs from FDMA as one channel occupies the entire
bandwidth of the link.
» CDMA differs from TDMA as all stations can send data
simultaneously; there is no time sharing.

» CDMA is based on CODING THEORY.

» Each station is assigned a code, which is a sequence of numbers,
called chips.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Note:

In CDMA, one channel carries all

transmissions simultaneously.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.14 Chip sequences

» Sequence of chips for stations A,B,C and D.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.15 Encoding rules

» If a station wants to send a 0 it sends a –1.

» If a station wants to send a 1 it sends a +1.
» If a station has nothing to send, or it is idle, this is represented by 0.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Multiplexer

» We have 4 stations that sharing the link in 1-bit interval.

» The procedure can easily be repeated for additional intervals.
» Station 1 and 2 are sending 0 bit and channel 4 is sending a 1 bit.
» Station 3 is silent.

» The multiplexer receives on encoded number from each station

(-1,-1,0,+1).
» The encoded number by 1 is multiplied by each chip in sequence A.
» A new sequence is in result as (-1,-1,-1,-1), similarly to station
2,3 and 4.
» All first chips are added, as all second, third, and fourth chips, the
result is a new sequence.
» The sequence is transmitted through the link.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.16 CDMA multiplexer

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 De-multiplexer

» The de-multiplexer receives the sequence sent across the link.

» It multiplies the sequence by the code for each receiver, the
multiplication is done chip by chip.
» The chips is each sequence are added. The result is always
+4,-4, or 0.
» The result is divided 4 to get -1,+1 or 0.
» And thus decoded to 0 or 1, or silence by receiver.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.17 CDMA demultiplexer

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Orthogonal Sequences and their Generation

» The sequences are not chosen randomly, they are carefully selected.
» Theses sequences are called Orthogonal Sequences.

» To generate the orthogonal sequences Walsh Tables are used.

» Walsh able is a 2-Dimensional table with an equal number of rows and
columns.
» Each row is a sequence of chips.
» The wash table W1 for a one-chip based sequence has 1 row and 1 column.
» We can choose –1 or +1 for the chip for this trivial table.
» According to Walsh if we the table for N sequences W N, we can create the
table for 2N sequences W2N.

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.18 Walsh table generation example:

W1 and W2N

» WN with the overhead bar stands for the complement of W N

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Figure 13.19 Sequence generation

» W2 can be made from FOUR W1’s, with last one complement.

» W4 can be made of 4 W2’s with the last one complemented
and so on…

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Properties of Orthogonal Sequences

» If we multiply the sequence by –1, every sequence is complemented,

if a station is sending –1 (bit 0), it is sending its complement.

» If we multiply two sequences, element by element and add the result
we get a number called the inner product.
» If the two sequences are same we get N, where N is the number of
sequences.
» If they are different we get 0.
» The inner product uses a dot as operator. So A.A=N but A.B=0.
» The inner product of a sequence by its complement is –N,
so A. (-A)= -N

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Example 1
Check to see if the second property about orthogonal codes holds
for our CDMA example.

Solution
The inner product of each code by itself is N. This is shown for code C; you
can prove for yourself that it holds true for the other codes.

C. C=


If two sequences are different, the inner product is 0.

B . C = 

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

 Example 2
Check to see if the third property about orthogonal codes holds for
our CDMA example.

Solution
The inner product of each code by its complement is N. This is shown for
code C; you can prove for yourself that it holds true for the other codes.

C . (C ) =


The inner product of a code with the complement of another code is 0.

B . (C ) =

McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004