EEE 497: Telecommunication Networks

Layering Architecture and MAC Protocols

Dr. Md. Forkan Uddin
Professor, Dept. of EEE, BUET, Dhaka-1205

1
 Protocol Hierarchies
 To reduce design complexity, most networks are organized as a
stack of layers or levels
Each one built upon the one below it
 The number of layers, the name of each layer, the contents of
each layer, and the function of each layer differ from network to
network
 The purpose of each layer is to offer certain services to the
higher layers
In a sense, each layer is a kind of virtual machine, offering
certain services to the layer above it

2
 OSI Layering Architecture/ Reference Model
 International standard organization (ISO) established a committee in 1977
to develop an architecture for computer communication
 Open Systems Interconnection (OSI) model is the result of this effort
 In 1984, OSI model is approved as reference model

3
 Upper Layers
Application Layer
 Allows access to network resource
 Make applications into data format
Presentation Layer
 Translate, encrypt and compress data
Session Layer
 The session layer is responsible for dialog
control and synchronization
Transport Layer
 Controls congestion
 Provides transmission reliability between source and destination
4
 Network Layer

The network layer is responsible for the delivery of

individual packets from the source host to the
destination host
 Determine the route for the packets

5
 Data Link Layer

 Who will transmit, when to transmit, whom to transmit to

 Attempts to provide reliable communication over the
physical layer interface in one hop distance

6 6
 Physical layer

 Provides physical interface

for transmission of information
through a medium (wired/wireless)

Covers all - mechanical, electrical, functional and procedural -

aspects for physical communication

7
Interaction among the layers

8 8
 Advantages/disadvantages of Layering
 Layer architecture simplifies the communication network design
 It is easy to debug network applications in a layered architecture
network
 The communication system management is easier
 Research/work on a layer can be done independently
 Some cases, the functionalities of multiple layer are dependent
especially for wireless networks
 Performance can be poor if they work independently
 Information from one layer need to pass to another layer and
the activity of a layer should be modified with those information
 Cross-layer design cross-layering of two or more layers and
optimization are proposing to overcome the problems
 Cross-layer examples:
MAC+PHY, Network+PHY, Network+MAC+PHY, Transport+MAC
9
 TCP/IP Reference Model
 All the OSI layers are not present

10 10
 TCP/IP Reference Model
TCP: Transmission Control Protocol
UDP: User Datagram Protocol
IP: Internet Protocol
ICMP : Internet Control Message Protocol
HTTP : HyperText Transfer Protocol
SMTP: Simple Mail Transfer Protocol
DNS: Domain Name System
RTP: Real time protocol

11 11
 What is MAC?

Multiple users under a single base station

Multiple Access (MA):
Decides on - Who will transmit? Whom to transmit? When to transmit?
How to transmit?
 MA techniques are channel access methods based on some
principles including multiplexing
 Allocates channels to different users and also handles the situation
when there are more message sources than available channels
 It works on the data link layer (L2) of OSI model

12
 Classification

Random access (contention methods): No station is superior to another station and none is
assigned the control over another. No station permits, or does not permit, another station to
send. distributed, low cost, poor performance
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. They were
designed for computer networks. Not popular now.
Channelization techniques: The available bandwidth of a link is shared in time, frequency, or
through code, between different stations. Usually, it is controlled by a system administrator.
Conflict free; centralized, high cost, good performance
 Frequency Division Multiple Access (FDMA)

 The frequency spectrum is divided into unique frequency

bands or channels
 These channels are assigned to users on demand
 Multiple users cannot share a channel at the same time
 In FDMA, the available bandwidth of the common channel is
divided into bands that are separated by guard bands
 FDMA requires tight RF filtering to reduce adjacent channel
interference

14
14
FDMA
 Time Division Multiple Access (TDMA)
 Time is slotted
 One user per slot
 Buffer and burst method
 Non-continuous transmission
 Digital data
 Digital modulation

16
16
TDMA
Code Division Multiple Access (CDMA)
 Transmitters may transmit at the same time and the same
channel
 Each signal is modified by spreading it over a large bandwidth

18
18
CDMA

19 19
CDMA

20
20
CDMA

21
21
 FDMA vs. TDMA vs. CDMA
FDMA
 Many channels - same antenna
 For maximum power efficiency amplifiers operate near saturation
 Near saturation power amplifiers are nonlinear
 Nonlinearities causes signal spreading
 Inter modulation (IM) frequencies
 IM are undesired harmonics
 Interference with other channels in the FDMA system
 Decreases user carrier to interference ratio (C/I) - decreases performance
 Interference outside the mobile radio band: adjacent-channel interference
 FDMA requires RF filtering to minimize adjacent channel interference
 RF filters needed - higher costs
 Even through, it could be a good choice for high speed communications
 FDMA vs. TDMA vs. CDMA
TDMA:
 Digital equipments used in time division multiplexing is increasingly becoming cheaper
 There are advantages in digital transmission techniques. Ex: error correction
 Lack of inter-modulation noise means increased efficiency
 Handoff process much simpler
 Low battery consumption due to sleep scheduling facility
 Extra signaling bits required for synchronization and framing
 High synchronization overhead and guard slots
 Selected for next generation communications
CDMA:
Near far problem in CDMA

 Requires power control

 Highly cost for power control circuit
 Good if the distance of the users are very close from the base station
Demand Assignment Multiple Access (DAMA)
 DAMA systems assign communication channels based on demand
from the clients
 When the channel is no longer in use, the channels are again returned
to the central pool for reassignment to other users
 DAMA is a protocol used in satellite communications, particularly Very
Small Aperture Terminal (VSAT) systems
 DAMA enables efficient and instantaneous assignment of transponder
channels on a first come, first served basis according to data priority
 ALOHA
 Each nodes access the channel distributed fashion
 When a packet arrive a node, it tries to access the channel and sends the packet the
receiver
 When multiple users access simultaneously collision occurs

2
ALOHA: Vulnerable time
 Throughput Model
 Let G is the traffic arrivals per packet time T

 Arrival rate λ=G/T

 In a frame time T, the probability of k transmission

Pr(N=k)=(GT/T)^k*exp(-GT/T)/k!

 Vulnerable period is 2T, where T is duration of transmission of a packet

 Probability of 0 transmission in 2T, i.e., probability of successful

transmission=exp(-2G)

 Throughput per packet duration, S=G*exp(-2G)

 Maximum throughput= 0.186, it means efficiency is 18.6%

 Slotted ALOHA
 Time is slotted
 Duration of each slot is equal to T
 Each nodes access the channel in a distributed fashion
 If a packet arrives to a node, it tries to access the channel in the next slot and
sends the packet the receiver
Slotted ALOHA: Vulnerable time
 Throughput Model
 Vulnerable period is T
 Probability of 0 transmission in T, i.e., probability of
successful transmission=exp(-G)

 Throughput per packet duration, S=G*exp(-G)

 Maximum throughput, S(max)= 0.368 , it means efficiency
is 36.8%
 ALOHA vs. Slotted ALOHA
Slotted ALOHA: the maximum throughput
Smax = 0.368=1/e when G = 1.

ALOHA: the maximum throughput

Smax = 0.184=1/2e when G = 0.5.
 CSMA
 In ALOHA and slotted ALOHA, a node does not have any knowledge, whether or not
the other node(s) are accessing the channel as a result collision is huge
 In CSMA, a node sense the channel by energy detection of the channel before a
transmission
 However, collision can happen due to the propagation delay and detection time
 Types of CSMA
 1-persistent: Each node senses
the channel continuously and
send whenever the channel is
idle
 Non-persistent: Senses the
channel in a time interval
 P-persistent: Senses
continuously but decision of how
many time slots after it will
transmit is based on random
probability distribution
 Types of CSMA
 1-persistent: collision is very high since more than one node
can start transmission at the same time

 Non-persistent: efficiency is better than 1-persistent

 P-persistent: efficiency is very good with appropriate selection

of p
 ALOHA vs. CSMA
0.01-persistent CSMA
1.0
0.9 Nonpersistent CSMA
0.8
0.7 0.1-persistent CSMA
0.6
0.5-persistent CSMA
0.5
S

1-persistent CSMA
0.4
0.3
0.2
Slotted Aloha
Aloha
0.1
0
0 1 2 3 4 5 6 7 8 9
G
 CSMA with collision detection
CSMA/CD
 In CSMA, if 2 terminals begin sending packet at the same time,
each will transmit its complete packet (although collision is
taking place).
 Wasting medium for an entire packet time.
 CSMA/CD is used to detect collision and stop transmission
 Used in Ethernet, very efficient
 Steps in CDMA/CD
Step 1: If the medium is idle, transmit
Step 2: If the medium is busy, continue to listen until the channel is idle then
transmit
Step 3: If a collision is detected during transmission, cease transmitting
Step 4: Wait a random amount of time and repeats the same algorithm
 Throughput CDMA/CD

 2tprop time is required to find out whether the channel has

been captured
 Success probability of a transmission,
 Optimal value of p=1/n and maximum success probability

 Let time slot duration =2tprop

 Throughput CDMA/CD
 The number of mini -slots required for successful transmission
be X; X has following PDF
Pr(X=k)=(1-Psucc)k-1 Psucc
 The average value of X
E[ X ]   k Pr( X  k )
k 1

  k (1  Psucc ) k 1 Psucc
k 1

Psucc

[1  (1  Psucc )]2
 1 / Psucc  e

 Maximum throughput,
Throughput CDMA/CD
 CSMA/CA (CSMA with collision Avoidance)
 All terminals listen to the same medium as CSMA/CD.
 Terminal ready to transmit senses the medium.
 If medium is busy it waits until the end of current transmission.
 It again waits for an additional predetermined time period DIFS (Distributed
inter frame Space).
 Then picks up a random number of slots (the initial value of back-off
counter) within a contention window to wait before transmitting its frame.
 If there are transmissions by other terminals during this time period (back-
off time), the terminal freezes its counter.
 It resumes count down after other terminals finish transmission + DIFS. The
terminal can start its transmission when the counter reaches to zero.
 CSMA/CA

Node A’s frame Node B’s frame Node C’s frame

Delay: B
Delay: C Time
Nodes B & C sense
the medium
Nodes C starts
Nodes B resenses the medium transmitting.
and transmits its frame.
Node C freezes its counter.

Nodes C resenses the

medium and starts
decrementing its counter.
 CSMA/CA

DIFS Contention
window DIFS Contention window

Medium Busy Next Frame

Time
Defer access Slot

Backoff after defer

DIFS – Distributed Inter Frame Spacing

 CSMA/CA with ACK
 Immediate Acknowledgements from receiver upon
reception of data frame without any need for sensing the
medium
 ACK frame transmitted after time interval SIFS (Short
Inter-Frame Space) (SIFS < DIFS)
 Receiver transmits ACK without sensing the medium
 If ACK is lost, retransmission done
 Collision/Contention Resolution
 Collision/ contention resolution technique used in CSMA/CA is the heart
of the protocol
 Adaptive CW management for collision/contention resolution
 For a packet, the initial CW is randomly selected from 0 to (CW_min-1)
 For the k-th transmission of a packet in case of (k-1) consecutive failure,
the CW is randomly selected from 0 to 2(k-1)CW_min-1
 It’s also called binary exponential back-off (BEB) mechanism
 After a successful transmission, for the next packet, the CW is reset to
randomly selected from 0 to (CW_min-1)
 Thus, the mechanism automatically reduce the collision if number of
stations increases
 CSMA/CA/ACK
DIFS Time
Data
Source
SIFS
ACK
Destination
DIFS Contention window

Next Frame
Other
Defer access Backoff after defer

SIFS – Short Inter Frame Spacing

 CSMA/CA with RTS/CTS
DIFS SIFS
RTS Data Time
Source
SIFS SIFS

CTS ACK
Destination
DIFS
Contention window

Next Frame
Other

Defer access Backoff after defer

 CSMA/CA
 Basic CSMA/CA is widely using in WiFi
 Very efficient for WLAN with a low number of stations
 Poor performance for multi-hop due to hidden and exposed
terminal problem
Collision
A B C D

HTP
A B C D

Missing opportunity
ETP
 Multiple Access Techniques in use
System Multiple Access
Advanced Mobile Phone System (AMPS) FDMA/FDD
Global System for Mobile (GSM) TDMA/FDD
US Digital Cellular (USDC) TDMA/FDD
Digital European Cordless Telephone (DECT) FDMA/TDD
US Narrowband Spread Spectrum (IS-95) CDMA/FDD
Satellite Communication TDMA and FDMA
WiMax TDMA
WiFi CSMA/CA
Ethernet CSMA/VD
IEEE 802 Protocols for LAN and MAN
 802.3 for LAN with Ethernet
 802.11 (wireless LAN)
802.16 (wireless MAN)

802.3 for LAN with Ethernet

Two kinds of Ethernet exist: classic Ethernet and switched
Ethernet
 Classic Ethernet solves the multiple access problem using the
techniques
 Switched Ethernet uses switches to connect different computers
 Classic Ethernet is the original form and ran at rates from 3 to 10
Mbps
Switched Ethernet runs at 100, 1000, and 10,000 Mbps, in forms
called fast Ethernet, gigabit Ethernet, and 10 gigabit Ethernet
 In practice, only switched Ethernet is used nowadays
 Classical Ethernet
Classic Ethernet uses bus topology, i.e., a single long cable to
which all the computers were attached
 Classic Ethernet uses the 1-persistent CSMA/CD algorithm
 Switched Ethernet
 Simplest way to connect multiple PCs with a hub or a switch
 Uses star topology
 Types: Fast Ethernet, Gigabit Ethernet, 10 Gigabit Ethernet
 802.11 (Wireless LAN)
 Two modes: infrastructure and ad hoc
 WLAN/WiFi: : infrastructure mode
 Emergency networks: ad hoc mode
 802.11 (Wireless LAN)
 Two frequency bands: 2.4 and 5.1 GHz
 802.11 tries to avoid collisions with a protocol called CSMA/CA
 802.16 (Wireless MAN)
 To stimulate the market, IEEE formed a group to standardize a
broadband wireless metropolitan area network
Informally the technology is called WiMAX (Worldwide
Interoperability for Microwave Access
 802.16 (Wireless MAN)
Licensed spectrum is the expected case for WiMAX, typically
around 2.5 or 3.5 GHz
 Flexible scheme for dividing the channel between stations, called
OFDMA (Orthogonal Frequency Division Multiple Access)
 WiMAX allows both methods : TDD and FDD
 The MAC sub-layer is connection-oriented and point-to-
multipoint, which means that one base station communicates with
multiple subscriber stations