Chapter4 MediumAccessControlSublayer

Computer Networks, Fifth Edition by Andrew Tanenbaum and David Wetherall, Pearson Education-Prentice Hall, 2011
The Medium Access Control

Sublayer
Chapter 4
The Medium Access Control Sublayer
Network Links can be divided into:
1. Point-to-point connections
2. Broadcast channels
Point-to-Point connections were discussed
in chapter 2.
Broadcast Links and their Protocols will be
discussed next.
In any broadcast network, the key issue is
how to determine who gets to use the
channel when there is competition.

Example: teleconferencing.

Also know as: Multi-access channels or
random access channels.
The protocols belong to sublayer of the data
link layer called Medium Access Control
(MAC) sublayer.
Especially important in LANs and especially
in wireless communication.
WANs use point-to-point links with exception
of satellite links.

Channel Allocation Problem
Static channel allocation
Assumptions for dynamic allocation.
Static Channel Allocation
Traditionally capacity of the channel is split
among multiple competing users (e.g., TDM
or FDM).
Example: FM radio stations.
However, when the number of senders is
large and varying or the traffic is bursty FDM
presents some problems.
If the spectrum is cut up into N regions and
Fewer than N users are currently interested in
communicating, a large piece of valuable
spectrum will be wasted.
More than N users want to communicate
some of them will be denied permission for
lack of bandwidth.
Dividing the channel into constant number
of users of static sub channels is inherently
inefficient.
A static allocation is poor fit to most
computer systems, in which data traffic is
extremely bursty:
Peak traffic to mean traffic rations of 1000:1
are common.
Consequently most of the channels will be
idle most of the time.
Example:
Mean Time delay T,
Channel capacity C,
Average rate l frames/sec
Frames average length of 1/m bits.
=
1
l

C = 100 Mbps,
1/m = 10,000 bits
l = 5000 frames/sec
T = 200 msec
This result holds only when there is no
contention in the channel.
=
1
l

Divide a single channel into N independent
channels:
C/N = 100/N Mbps,
1/m = 10,000 bits
l/N = 5000 frames/sec
T
N
= Nx200 msec
For N=10 => T
N
= 2 msec.
=
1
=

l
= NT

Assumptions for Dynamic Channel
Allocation
1. Independent traffic
2. Single channel
3. Observable Collisions
4. Continuous or slotted time
5. Carrier sense or no carrier sense
Allocation
Independent Traffic:
The model consists of N independent
stations.
The expected number of frames generated in
an interval of length is . is arrival
rate of new frames.
Once the frame has been generated, the
station is blocked and does nothing until the
frame has been successfully transmitted.

Allocation
Single Channel:
The single channel is available for all
communication.
All stations can transmit on it and all can
receive from it.
The stations are assumed to be equally
capable though protocols may assign then
different roles (i.e., priorities)
Allocation
Observable Collisions:
If two frames are transmitted simultaneously,
they overlap in time and the resulting signal
is garbled.
This event is know as collision.
All stations can detect that a collision has
occurred. A collided frame must be
retransmitted.
No errors other than those generated by
collision occur.
Allocation
Continuous or Slotted Time:
Time may be assumed continuous. In which
case frame transmission can begin at any
instant.
Alternatively, time may be slotted or divided
into discrete intervals (called slots).
Frame transmission must hen begin at the
start of a slot.
A slot may contain 0, 1 or more frames,
corresponding to an idle slot, a succeful
transmission, or collision, respectively.
Allocation
Carrier Sense or No Carrier Sense:
With the carrier sense assumption, stations
can tell if the channel is in use before trying
got use it.
No station will attempt to use the channel
while it is sensed as busy.
If there is no carrier sense, stations cannot
sense the channel before trying to use it.
They will transmit then. One later they can
determine whether the transmission was
successful.
Allocation
Poisson models are used to model
independence assumption due to its
tractability. This is know to not be true.
Single channel assumption is the heart of
the model. This models is not a good model.

Multiple Access Protocols
ALOHA
Carrier Sense Multiple Access
Collision-free protocols
Limited-contention protocols
Wireless LAN protocols


ALOHA
1970 Hawaii
Norman Abramson and colleagues
have enabled wireless communication
between users in a remote island to the
central computer in Honolulu.
Two versions of the protocol now called
ALOHA:
Pure ALOHA and
Slotted ALOHA

Pure ALOHA
Each user is free to transmit whenever they have
data to be sent.
There will be collisions
Senders need some way to fond out if this is the case.
In ALOHA after the satiation transmits its message
to the central computer, the computer rebroadcast's
the frame to all of the stations.
Original sending station can listen for the broadcast
from the hub to see if its frame has gone through.
Pure ALOHA
In other wired systems the sender might be able to
listen for collisions while transmitting.
If the frame is destroyed, the sender just waits a
random amount of time and sends it again.
Waiting time must be random or the sending frames
will collide over and over.
Contention systems: that use the same channel in
the way that might lead to conflicts.


PURE ALOHA (1)
In pure ALOHA, frames are transmitted
at completely arbitrary times
Collision
Collision
Time
User

A

B

C

D

E
Pure ALOHA
What is the efficiency of an ALOHA channel?
Infinite collection of users typing at their terminals
(stations).
User states: WAITING or TYPING.
When a line is finished, the user stops typing waiting
for response.
The station then transmits a frame containing the
line over the shared channel to the central computer
and checks the channel to see if it was successful.
If so the users sees the reply and goes back to typing
If not, the user continuously to wait while the station
retransmits the frame over and over until it has been
successfully send.

Pure ALOHA
Frame Time denotes the amount of time needed
to transmit the standard, fixed-length frame.
Each new frame is assumed to be generated by
Poisson distribution with a mean of N frames per
frame time.
If N>1 the user community is generating frames at a
higher rate than the channel can handle, and nearly
every farm will suffer a collision.
For reasonable throughput we expect 0 < N < 1.
Pure ALOHA
In addition to the new frames, the stations also
generate retransmissions of frames that previously
suffered collisions.
Assume that the new and the old frames combined
are well modeled by a Poisson distribution with
mean G frames per frame time. .
Low load: 0 there will be few collisions, hence
few retransmissions,
High load: there will be many collisions, > .

Pure ALOHA
Under all loads the throughput S is just the offered
load, G, times the probability P
0
of a transmission
succeeding:
=
0


ALOHA (2)
Vulnerable period for the shaded frame.

Pure ALOHA
The probability that k frames are generated
during a given frame time, in which G frames are
expected, is given by the Poison distribution:
Pr [] =

!

Probability of zero frames:

In an interval two frame times long, the mean
number of frames generated is 2G.
Probability of no frames being initiated during
the entire vulnerable period is given by
0
=
2
.
Using =
0


Pure ALOHA
=
2
.

The relation between the offered traffic
and the throughput is given in the next
slide.
The maximum throughput occurs at G=0.5
with S=1/2e which is about 0.184.
The maximum utilization of the channel thus is 18%.


ALOHA (3)
Throughput versus offered traffic for ALOHA systems.
Sloted ALOHA
Roberts in 1972 doubled the capacity of an
ALOHA system.
Divide time into discrete intervals called slots.
Each interval corresponds to one frame.
Users will have to agree on slot boundaries.
Synchronization is required:
One special station emit a pip at the start of
each interval, like clock.
Sloted ALOHA
A station is not permitted to send whenever
the user types a line.
User waits for the beginning of the next slot.
Continuous time ALOHA is turned into a
discrete time one.
The probability of no other traffic during the
same slot as our test frame is then
,
which leads to:
=

Slotted ALOHA
Slotted ALOHA
peaks at the G = 1
Throughput S = 1/e = 0.367 or 37%.
The best case scenario:
37% of slots are empty
37% of successes, and
26% collisions.

Carrier Sense Multiple Access
Protocols
Protocols in which stations listen for a
carrier (i.e., transmission) and act
accordingly are called carrier sense
protocols.
Several Versions of those protocols will be
discussed.
1. Persistent and Nonpersistent CSMA
2. CSMA with Collision Detections

Persistent and Nonpersistent CSMA
1-Persistend Carrier Sense Multiple
Access (CSMA) protocol.
When a station has data to be send it first
listens to the channel to see if anyone else is
transmitting at that moment.
If the channel is idle the station sends the
data,
Otherwise, the station just waits until it
becomes idle.

1-Persistend Carrier Sense Multiple
Access (CSMA) protocol.
If a collision occurs, the station waits a
random amount of time and starts all over
again.

This protocol has problems with collisions:
2 patiently waiting stations will start transmitting
at the same time when the channel becomes idle.
Propagation delay can make even more suble the
collision.
1-persistent refers to the probability of 1 of
transmission when the channel if found to
be idle.

Nonpersistent CSMA - Second Carrier
Sense protocol is. In this protocol the
transmitting stations are less greedy.
The transmitting station will send the packet
if the channel is found to be idle, however
If the channel is already in use the station
does not continuously sense it for
transmission. Instead it waits a random
amount of time and then repeats the
algorithm.
P-persistent CSMA.
The transmitting station will send the packet
if the channel is found to be idle with a
probability of p (q = 1-p; it defers that action
until the next slot).
If the slot is still empty it does or not transmit
with the probability of p and q respectively.
If the channel in use the station will treat this
as being a collision (waits random amount of
time)

Comparison of the channel utilization versus load for various
random access protocols.
CSMA with Collision Detection
Protocols that sense Collisions are know as
(CSMA/CD)
This protocol is a basis of classical Ethernet
LAN.
The transmitting station is reading the data
that it is transmitting.
If it is garbled up then it will know that
collision has occurred.

CSMA/CD can be in one of three states: contention,
transmission, or idle.
In CSMA/CD collisions do not occur once
the station has unambiguously captured the
channel, but they still occur during the
contention period.
These collisions adversely affect the system
performance (e.g., bandwidth-delay product
is large long cable that has a large
propagation delay t and frames are short).
Collision-Free Protocols
Collisions reduce the bandwidth
The increase the time to send a frame
Bad fit for real-time traffic:
VoIP
Video,
Teleconferencing, etc.
Collision-Free Protocols
N Stations
Each programmed with a unique address:
0-(N-1).
Propagation delay we assume to be
negligible.

Question: Which station gets the channel
(e.g., the right to transmit) after a successful
transmission.
Basic Bit-Map Protocol
Each contention period consists of exactly N slots.
If station 0 has a frame to send, it transmits a 1 bit
during the slot 0.
No other station is allowed transmit during this slot.
Regardless what station 0 does, station 1 gets to
opportunity to transmit a 1 bit during slot 1, but only
if it has a frame queued.
In general, station j may announce that it has a
frame to send by inserting a 1 bit into slot j.
After all N slots have passed by, each station has
complete knowledge of which stations wish to
transmit. At which point they begin transmitting
frames in numerical order.


Collision-Free Protocols (1)
The basic bit-map protocol.
Bit-Map Protocol
Protocols that broadcast their intention before that
actually transmit are called reservation protocols.
Low-load conditions:
Average wait conditions for low-numbered stations:
N/2 slots for current scan to finish, and
N slots for the following scan to run to completion before it
may begin transmitting.
1.5N slots wait time.
Average wait conditions for high-numbered stations:
0.5N slots wait time.
Mean of all stations is N times.

Bit-Map Protocol
Efficiency:
Overhead bits N
Data bits d
High-load
N bit contention period is prorated over N frames,
yielding an overhead of only 1 bit per frame:
Efficiency:

Mean delay:
Sum of the time it queues in the station +
(N-1)d + N

+1

+

Token Passing
Message is passed called token form station to the
next in the same predefined order.
Token Ring or Token Bus protocols work the same
way.
One has to pay attention to the ring because if it is
not removed from circulation it will end up being
there forever.
Typically it will be removed by the receiving station
and/or sending station.


Collision-Free Protocols (2)
Token ring.
Station
Direction of
transmission
Token

Binary Countdown
A problem with the basic bit-map and token passing
protocols is the overhead of 1 bit per station.
Large overhead for the network with large number of
stations.
A better solution is to use binary station addresses
with a channel that combines transmissions.
A station wanting to use the channel now
broadcasts its address as a binary bit string,
starting with the high-order bit. The addresses are
assumed to be the same length.
The bits in each address position from different
stations are BOOLEAN.
The are OR-ed together by the channel when they are
send at the same time.
Binary Countdown protocol

Binary Countdown
Arbitration rule: As soon as a station sees that a
high-ordered bit position that is 0 in its address has
been overwritten with 1 it gives up.
Example:
If stations 0010, 0100, 1001, and 1010 are all trying
to get the channel, in the first bit time the stations
transmit 0, 0, 1, and 1, respectively.
They are OR-ed together to get 1.
Stations 0010 and 0100 see the 1 and know that
higher-numbered stations is competing for the
channel and they give up for the current round.
Stations 1001 and 1010 continue.
Binary Countdown
The next bit is 0 so both stations continue.
The next bit is 1 so the station 1001 gives up and
station 1010 wins the bidding.
This gives it a right to transmit the frame, after which
a new cycle starts.


Binary Countdown
The binary countdown protocol. A dash indicates silence.
Limited-Contention Protocols
So far we have considered two basic strategies for
channel acquisition in a broadcast network:
Contention (e.g., CSMA), and
Collision free protocols.
Two important performance measures:
Delay at low-loads, and
Channel efficiency at high-loads.
Pure or Slotted ALOHA is preferred under
The low load conditions:
Low delay and
practically collision free.
The high load conditions:
High Delay due to
High number of collisions or contentions
Reverse is true for collision-free protocols
The low load conditions:
High delay and
The high load conditions:
Relatively low Delay due to,
Channel efficiency improves (fixed overheads).

Symmetric Limited-Contention Protocol
k stations are contending for channel access.
Each station has p probability of transmitting
during each slot
Probability that any station acquires a channel is its
probability p multiplied with all the remaining (k-1)
stations differing with probability of (1-p):
1
1

This probability is displayed in the next slide.

Acquisition probability for a symmetric contention channel.
From the figure in previous slide clearly that probability
that some stations will acquire the channel can be
increased only by decreasing the amount of
competition.
The limited contention protocols do just that by:
1. Dividing the stations into (not necessarily disjoint)
groups.
2. Only the members of group 0 are permitted to compete
for slot 0.
3. If one of them succeeds, it acquires the channel and
transmits its frame.
4. If there is a collision the members of the group 1
contend for slot 1. etc.
By making an appropriate division of stations into
groups the amount of contention for each slot can be
reduced, thus operating each slot near the left of the
figure presented in previous slide.

The trick is how to assign stations to slots?
Such assignment guarantees that there will never be
collisions because at most one station is contending for
any given slots (binary countdown protocol)
The next case us to assign two stations per group. The
probability that both will try to transmit during a slot is p
2

which for small p is negligible.
We need a way to assign station slots dynamically:
Many stations per slot when the load is low, and
Few (or just one) station per slot when the load is high.
The Adaptive Tree Walk Protocol
Algorithm used for testing soldiers during World
War II:
Blood samples from N soldiers
A portion of each sample was poured into a single
test tube.
If this mixed sample was testing:
If none of antibodies were found all the soldiers in the
group were declared healthy.
Binary search was performed to pick which soldier was
infected.


The tree for eight stations
Slot 0 - First contention slot following the successful
transmission when all stations were permitted to try
to acquire the channel.
Slot 1 If there is a collision then during slot 1 only
those stations falling under node 2 in the tree (next
slide) may compete.
Slot 2 - If one of them acquires the channel the slot
following the frame is reserved for those stations
under node 3. If on the other hand two or more
stations under node 2 want to transmit, there will be
a collision during slot 1, in which case it is node 4s
turn.

Depth first search of the tree to locate all ready
stations if the collision occurs during slot 0. Each bit
slot is associated with some particular node in the
tree.
If collision occurs the search continues recursively
with the nodes left and right children.
If a bit slot is idle or if only one station transmits in
it. The searching of its node can stop because all
ready stations shave been located.
When a load on the system is high it is not
worth to dedicate slot 0 to node 1.
Similarly one would argue that nodes 2 and
3 should be skipped.
In general the question is at what level in
the tree should we began the search?
Heavier load the farther down the tree the
search should begin.

We will assume that each station has a
pretty good idea of the number of ready
stations q, (from monitoring traffic).
Numbering the levels:
Level 0: Node 1
Level 1: Nodes 2, 3
Level 2: Nodes 4,5,6 and 7. etc.

q ready stations are uniformly distributed.
Expected number of the stations below a
specific node at level i is just 2
-i
q
Optimal number of contending station per slot
should be 1 and hence 2
-i
q = 1.
Hence:
=
2

Wireless LAN Protocols
A system of laptop computers that
communicate by radio wireless LAN.
It also has somewhat different properties
than a wired LAN.
Leads to different MAC protocols.

Common configuration of wireless LAN:
Office Building with Access Points (APs)
APs Strategically placed
APs are wired together (copper or fiber)
APs provide connectivity

Transmission power of APs and laptops is
adjusted to have a range of tens of meters
nearby rooms becomes like a single cell and
the entire building becomes like the cellular
telephony system.
Each cell has only one channel.
This channel is shared by all the stations in
the cell, including APs.
Bandwidth provided up to 600 Mbps.

Wireless system can not normally detect a
collision while it is occurring.
The received signal is weak (millions times
fainter than the signal that is being
transmitted)
Difficulty in finding it.
Instead ACK are used to discover collisions
and other errors after the fact.
Additional, and even more important,
difference between wireless LAN and wired
LAN:
Wireless LAN: A station may not be able to
transmit or receive frames to or from all
other stations due to limited radio range.
Wired LAN: Once the one station sends a
frame, all other stations receive it.
Simplifying assumptions:
Each radio transmitter has some fixed range.
Its range is represented by an ideal circular
coverage region
within that region station can sense and
receive the stations transmission.

Nave approach:
Use CSMA:
Just listen for other transmissions.
In none is doing it then transmit.
Problem:
What matters for reception is interference at the
receiver and not at the sender.

See figure in next slide:

Wireless LAN Protocols (1)
A wireless LAN. (a) A and C are hidden terminals
when transmitting to B.


A wireless LAN. (b) B and C are exposed terminals when
transmitting to A and D.
Before starting the transmission a station must
know whether there is radio activity around the
receiver.
CSMA merely tells it whether there is activity near
the transmitter by sensing the carrier.
With wired communication all signals propagate to
all stations so this distinction does not exist.
However only one transmission can take place at one
time.
In a system based on short-range radio waves
multiple transmissions can occur simultaneously:
If they all have different destinations, and
These destinations are out of range of one another.
Multiple Access with Collision Avoidance (MACA) -
Early protocol that tackles these problems for
wireless LANs.
The basic idea behind it is for the sender to simulate
the receiver into outputting a short frame
Nearby stations can detect this transmission and
avoid transmitting for the duration of the upcoming
(larger) data frame.
In a system based on short-range radio waves
multiple transmissions can occur simultaneously:
If they all have different destinations, and
These destinations are out of range of one another.
Multiple Access with Collision Avoidance
(MACA) - Early protocol that tackles these problems
for wireless LANs.
The basic idea behind it is for the sender to simulate
the receiver into outputting a short frame
Nearby stations can detect this transmission and
avoid transmitting for the duration of the upcoming
(larger) data frame.
MACA is illustrated in the next slide.
A sends a frame to B A initiates the request by
sending an Request To Send (RTS) to station B.
Short frame (30 bytes) that contains the length of data
frame that will eventually follow.
B replies with a Clear To Send (CTS) frame.
This frame contains the data length (copied from RTS).
After reception of the CTS frame the a station A
begins transmission.

The MACA protocol. (a) A sending an RTS to B. (b) B
responding with a CTS to A.
Any station hearing the RTS is clearly close to A
and must remain silent long enough for the CTS to
be transmitted back to A without conflict.
Any stations hearing CTS are clearly close to B and
must remain silent during the upcoming data
transmission, whole length it can tell by examining
the CTS frame.
Collisions are possible.

Ethernet
Physical layer
MAC sublayer protocol
Ethernet performance
Switched Ethernet
Fast Ethernet
Gigabit Ethernet
10 Gigabit Ethernet
IEEE 802.2: Logical Link Control
Retrospective on Ethernet

Ethernet
Classical Ethernet
Switched Ethernet
Classical Ethernet
Bob Metcalfe with David Boggs designed
and implemented the first local area network
in 1976 in Xerox Palo Alto Lab.
It used a ingle long thick coaxial cable.
Speed 3 Mbps.
Ethernet luminiferous ether.
Successful designed that was later drafted
as standard in 1978 by Xerox, DEC, Intel
with a 10 Mbps.
In 1983 it became the IEEE 802.3 standard
Classical Ethernet
Thick Ethernet a thick cable. Segment could be
as long as 500 m. Could be used to connect up to
100 computers.
Thin Ethernet BNC connectors. Segment could be
no longer than 185 m. Could be used to connect up
to 30 computers.
For a large length connectivity the cables could be
connected by repeaters.
Repeater is a physical layer device that receives,
amplifies, and retransmits signals in both directions.

Classic Ethernet Physical Layer
Architecture of classic Ethernet

Classical Ethernet
Over each of those cables the signal was coded
using Manchester encoding.
Other restriction was that no two transceivers
could be more than 2.5 km apart and no path
between any two transceivers could traverse
more than four repeaters.
This limitation was impose due to the MAC
protocol used.

MAC Sublayer Protocol
The format used to send frames is shown in
the figure in next slide.
MAC Sublayer Protocol (1)
Frame formats. (a) Ethernet (DIX). (b) IEEE
802.3.
Classic Ethernet MAC Sublayer
Protocol
Format to send frames is shown in the figure in the
previous slide.

1. Preamble 8 bytes
7x 10101010 and 10101011 <- Start of Frame
Delimiter (802.3).
The Manchester encoding of this pattern produces
10-MHz wave for 6.4 msec used for
synchronization.
The last two bits indicate the start of the frame.

Classic Ethernet MAC Sublayer
Protocol
2. Two addresses each 6 bytes destination +
source
First bit of the destination address is 0 for ordinary addresses
and 1 for group addresses.
Group address allow multiple destinations to listen to a single
address Multicasting.
Special address consisting of all 1 is reserved for broadcasting.
Uniqueness of the addresses:
First 3 bytes are used for (Organizationally Unique Identifier)
Blocks of 2
24
addresses are assigned to a manufacturer.
Manufacturer assigns the last 3 bytes of the address and
programs the complete address into the NIC.

3. Type or Length field.
Depending whether the frame is Ethernet or IEEE
802.3
Ethernet uses a Type field to tell the receiver what
to do with the frame.
Multiple network-layer protocols may be in use at
the same time on the same machine. So when
Ethernet frame arrives, the operating system has to
know which one to hand the frame to. The Type
field specifies which process to give the frame to.
E.g. 0x0800 indicates the frame contains IPv4
packet.
3. Type or Length field.
Depending whether the frame is Ethernet or IEEE
802.3
Ethernet uses a Type field to tell the receiver what
to do with the frame.
Multiple network-layer protocols may be in use at
the same time on the same machine. So when
Ethernet frame arrives, the operating system has to
know which one to hand the frame to. The Type
field specifies which process to give the frame to.
E.g. 0x0800 indicates the frame contains IPv4
packet.
Length of the field could be carried as well.
Ethernet length was determined by looking inside
the data a layer violation.
Added another header for the Logical Link Control
(LLC) protocol within the data. It uses 8 bytes to
convey the 2 bytes of protocol type information.
Rule: Any number greater than 0x600 can be
interpreted a Type otherwise is considered to be
Length.
4. Data Field
Up to 1500 bytes.
Minimum frame length valid frames must be at
least 64 bytes long from destination address to
checksum.
If data portion is less than 46 bytes the Pad field is
used to fill out the frame to the minimum size.
Minimum filed length is also serves one very
important role prevents the sender to complete
transmission before the first bit arrives at the
destination.

MAC Sublayer Protocol (2)
Collision detection can take as long as 2t.
10 Mbps LAN with a maximum length of 2500 m
and four repeaters the round-trip time has been
determined to be nearly 50 msec in the worst case.
Shortest allowed frame must take at least this long
to transmit.
At 10 Mbps a bit takes 100 nsec
500 bits (numbit = 10 Mbps X 100 nsec) rounded up
to 512 bits = 64 bytes.
4. Checksum
It is a 32-bit CRC of the kind that we have covered
earlier.
Defined as a generator polynomial described in the
textbook.
Ethernet Performance
Each station transmits during a contention
slot with probability p.
The probability that some station acquires
the channel A:

= 1
1

Max A for p=1/k with A .

The probability that contention interval has
exactly j slots in it is A(1-A)
j-1
.
Mean number of slots per contention is:

1
1
=0
=
1

Duration of each slot is 2t, the mean
contention interval w is = 2t/A

Assuming optimal p, the mean number of
contention slots is never more than e, thus
w is at most 2te 5.4t.

If the mean frame takes P sec to transmit, when
many stations have frames to send channel
efficiency E
=
+2/

Here we see where the maximum cable distanced
between any two stations enters into the
performance figures.
The longer the cable the longer the contention
interval; This is why the Ethernet standard specifies
the maximum cable length.
It would be instructive to reformulate the equation in
the previous slide in term of the frame length F,
network bandwidth B and the cable length L, speed
of signal propagation c, for the optimal case e
contention slots per frame.
P = F/B the equation becomes:
=
1
1 +2/

When the term 2/ >> 0 the network efficiency
becomes very low.
Increasing BL; Bandwidth and/or Length of the cable
reduces the efficiency.
This is contrary to the design criteria to have largest
possible bandwidth and longest connections.
Classical Ethernet will not be able to provide this.


Efficiency of Ethernet at 10 Mbps with 512-bit slot times.
The theoretical result that Ethernet can not
work that efficiently is flowed due to several
reasons:
Poison distribution of the traffic is not
realistic.
Research focuses on only several
interesting cases.
Practical results show otherwise that the
Ethernet works.

Switched Ethernet
Wiring was changed from a long cable architecture
to a more complex architecture:
Each station has a dedicated cable running to a
central hub. (Fig (a) in the next slide).
Adding and removing a station become much easier.
Cable length was reduced to 100 m for telephone
twisted pair wires and to 200 hundred if Category 5
cable was used.
Hubs do not increase capacity they are equivalent
to the single long cable of classic Ethernet.
As more stations were added the performance of each
station degraded.


Switched Ethernet (1)
(a) Hub. (b) Switch.
Switched Ethernet
One could solve this problem by increasing the
speech of the basic Ethernet from 1 Mbps to 10
Mbps, 100 Mbps or even 1 Gbps.
However, multimedia applications requires even
higher bandwidths.
Switch is the solution.
Switch must be able to determine which frame goes
to what station.
Security benefits
No collision can occur.

Switched Ethernet (2)
An Ethernet switch.
Switch
Twisted pair
Switch ports
Hub

Fast Ethernet
The original fast Ethernet cabling.
GigaBit Ethernet
After the standard for Fast Ethernet was adopted the
work for yet even faster standard started: GigaBit
Ethernet
Goals:
Increase performance ten fold over Fast Ethernet.
Maintain compatibility with both Classical and Fast
Ethernet.
Unacknowledged datagram service with both unicast and
broadcast.
Use the same 48-bit addressing scheme already in use,
Maintain the same frame format including minimum and
maximum sizes.


Gigabit Ethernet (1)
A two-station Ethernet

A two-station Ethernet

Gigabit Ethernet cabling.

10 Gigabit Ethernet
Gigabit Ethernet cabling

Wireless Lans
802.11 architecture and protocol stack
802.11 physical layer
802.11 MAC sublayer protocol
802.11 frame structure
Services


802.11 Architecture and Protocol Stack (1)
802.11 architecture infrastructure mode
Access
Point
Client
To Network

802.11 architecture ad-hoc mode

Part of the 802.11 protocol stack.

The 802.11 MAC Sublayer Protocol (1)
Sending a frame with CSMA/CA.

The hidden terminal problem.

The exposed terminal problem.

The use of virtual channel sensing using CSMA/CA.

Interframe spacing in 802.11

802.11 Frame Structure
Format of the 802.11 data frame

Broadband Wireless
Comparison of 802.16 with 802.11, 3G
802.16 architecture and protocol stack
802.16 physical layer
802.16 frame structure


Comparison of 802.16 with 802.11 and 3G
The 802.16 architecture

802.16 Architecture and Protocol Stack
The 802.16 protocol stack

802.16 Physical Layer
Frames structure for OFDMA with time division duplexing.

802.16 MAC Sublayer Protocol
Classes of service

1. Constant bit rate service.
2. Real-time variable bit rate service.
3. Non-real-time variable bit rate service.
4. Best-effort service.

802.16 Frame Structure
(a) A generic frame. (b) A bandwidth request frame.

Bluetooth
Architecture
Applications
Protocol stack
Radio layer
Link layers
Frame structure

Bluetooth Architecture
Two piconets can be connected to form a scatternet

Bluetooth Protocol Stack
The Bluetooth protocol architecture.

Bluetooth Frame Structure
Typical Bluetooth data frame at (a) basic, and
(b) enhanced, data rates.

RFID
EPC Gen 2 architecture
EPC Gen 2 physical layer
EPC Gen 2 tag identification layer
Tag identification message formats

EPC Gen 2 Architecture
RFID architecture.

EPC Gen 2 Physical Layer
Reader and tag backscatter signals.

EPC Gen 2 Tag Identification Layer
Example message exchange to identify a tag.

Tag Identification Message Formats
Format of the Query message.

Data Link Layer Switching
Uses of bridges
Learning bridges
Spanning tree bridges
Repeaters, hubs, bridges, switches,
routers, and gateways
Virtual LANs


Learning Bridges (1)
Bridge connecting two multidrop LANs

Bridges (and a hub) connecting seven point-to-point stations.

Protocol processing at a bridge.

Spanning Tree Bridges (1)
Bridges with two parallel links

Spanning Tree Bridges (2)
A spanning tree connecting five bridges. The dotted lines are
links that are not part of the spanning tree.

Poem by Radia Perlman (1985)
Algorithm for Spanning Tree (1)
I think that I shall never see
A graph more lovely than a tree.
A tree whose crucial property
Is loop-free connectivity.
A tree which must be sure to span.
So packets can reach every LAN.
. . .

Poem by Radia Perlman (1985)
Algorithm for Spanning Tree (2)
. . .
First the Root must be selected
By ID it is elected.
Least cost paths from Root are traced
In the tree these paths are placed.
A mesh is made by folks like me
Then bridges find a spanning tree.

Repeaters, Hubs, Bridges, Switches,
Routers, and Gateways
(a) Which device is in which layer.
(b) Frames, packets, and headers.

Virtual LANs (1)
A building with centralized wiring using hubs and a switch.

Virtual LANs (2)
Two VLANs, gray and white, on a bridged LAN.

The IEEE 802.1Q Standard (1)
Bridged LAN that is only partly VLAN-aware. The shaded
symbols are VLAN aware. The empty ones are not.

The IEEE 802.1Q Standard (2)
The 802.3 (legacy) and 802.1Q Ethernet frame formats.

End
Chapter 4

Chapter4 MediumAccessControlSublayer

Uploaded by

Copyright:

Available Formats

Chapter4 MediumAccessControlSublayer

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter4 MediumAccessControlSublayer

Uploaded by

Copyright:

Available Formats

Computer Networks, Fifth Edition by Andrew Tanenbaum and David Wetherall, Pearson Education-Prentice Hall, 2011

The Medium Access Control

You might also like