Internet Working

INTERNETWORKING
UTTAM K. ROY
Dept. of Information Technology,

Jadavpur University, Kolkata
Internetworking 2
Advanced Java Programming Books

• Andrew S. Tanenbaum, “Computer Networks”, Prentice Hall of India
publication
• William Stallings, “Data & Computer Communications”, Prentice Hall

of India publication
• Alberto Leon Garcia & Indra Widjaja, “Communication Networks”, Tata

McGraw-Hill publication
• Krouse, “Computer Networks”, Pearson publication
• Farouzan, “Computer Networks”
• Farouzan, “TCP/IP Protocol Suit”
© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Internetworking 3
Advanced Java Programming Introduction

 Definition
 Collection of (autonomous) computers connected by some
fashion
 Example

Internetworking 4
Advantages of Computer Networks

Advanced Java Programming
 Resource Sharing:
 Expensive resources (Database, high-speed laser printer,
array processors) can be shared among users of different
sites as they are connected to one another.
 Example
• World Wide Web(WWW)
• File Transfer Protocol(FTP)
• Domain Name System
• Network File System
• Database
• Remote login
 Computation/Productivity speedup:
 Concurrent execution of independent tasks using cluster
 Example
• Parallel Binary search
• Matrix multiplication
• Load balancing
Internetworking 5
Advantages of Computer Networks

 Reliability/Availability/Fault Tolerant:
 Entire system does not go down if a part becomes faulty
 Example
• Introduction of secondary DNS
• Distributed database
 (Inter-process)Communication:
 Processes running at different computers can communicate
 Example
• Remote Procedure Call
• Socket
 Incremental growth:
 Avoid huge initial setup cost
 Improved Control/Flexibility:
 Facility to control the system remotely
 Responsiveness:
Internetworking 6
Communications Model
 Source
 generates data to be transmitted—telephones, PC
 Transmitter
 Converts data into transmittable signals—Modem
 Transmission System
 Carries data
 Receiver
 Converts received signal into data
 Example
• Modem
 Destination
 Takes incoming data

Internetworking 7
Key Communications Tasks

 Transmission System Utilization

 Efficient use of shared medium
 Example
• Multiplexing—FDM, TDM, WDM
• Congestion control
 Interfacing
 Interfacing with transmission medium—electromagnetic
signal propagation
 Signal Generation
 Form and intensity such that
• Capable of being propagated
• Interpretable at receiver
 Synchronization
 Determining beginning and ending of signals and its duration

Internetworking 8

 Exchange Management
 Cooperation(Convention) between communicating parties
• Format and direction(simplex, duplex)
 Error detection and correction
 Signals are distorted before they reach at the destination
 Some tasks can not tolerate error—FTP
 Recovery
 Next step to the error detection
 Flow Control
 Fast transmitter/slow receiver problem
 Addressing and routing
 Identifying network devices

Internetworking 9

 Message formatting
 Agreement between two parties—binary code for characters
 Security
 Sender wish to be assured that intended receiver gets data
 Receiver must be sure that the data have not been altered
in transit
 Receiver must be sure that the data come from purported
sender not from intruder
 Network Management
 Configuration
 Monitor its status
 Responds on failure and overloads

Internetworking 10
Simplified Data Communications Model


Protocol
Architecture
Internetworking 12
Advanced Java Programming Protocols

 High degree of cooperation is needed between
two computer systems
 Used for communications between entities in a
system
 Must speak the same language
 Entities
 User applications
 e-mail facilities
 terminals
 Systems
 Computer
 Terminal
 Remote sensor

Internetworking 13
Advanced Java Programming Protocols

 Key elements of a Protocol
 Syntax
• Data formats
• Signal levels
 Semantics
• Control information
• Error handling
 Timing
• Speed matching
• Sequencing
 Standardized protocol
 Needed to promote interoperability among vendor
equipment

Internetworking 14
Advanced Java Programming OSI Model

 Open Systems Interconnection
 Developed by the International Organization
for Standardization (ISO) in 1977
 Seven layers
 Application
 Presentation
 Session
 Transport
 Network
 Data Link
 Physical
 A theoretical system delivered too late!
 TCP/IP is the de facto standard

Internetworking 15

 Physical Layer
 Physical interface between data transmission device (e.g.
computer) and transmission medium or network
 Characteristics of transmission medium
• Mechanical—connector type
• Electrical—signal levels
• Functional—function of individual cuircits
• Procedural—sequence of events,data rates etc.
 Data Link Layer
 Error detection and correction
 Flow control
 Network Access Layer
 Exchange of data between end system and network
 Destination address provision
 Routing functions across multiple networks

Internetworking 16

 Transport Layer
 Connection oriented and connectionless service
 Reliable delivery of data
 Sequencing/Ordering of delivery
 Avoid duplication
 Session Layer
 Dialog Discipline—full duplex or half duplex
 Recovery—check pointing mechanisms
 Presentation Later
 Format/presentation/syntax of data
 Application Layer
 Provides user interface such as file transfer (FTP), electronic
mail(SMTP), remote login(Telnet/SSH/rlogin), WWW(http)
etc.

Internetworking 17
TCP/IP Protocol Architecture

 Developed by the US Defense Advanced

Research Project Agency (DARPA) for its
packet switched network (ARPANET)
 Used by the global Internet
 No official model but a working one.
 Application layer
 Host to host or transport layer
 Internet layer
 Network access layer
 Physical layer

Internetworking 18
Advanced Java Programming OSI v TCP/IP

Internetworking 19
TCP/IP Protocol Architecture Model


Data
Transmission
Internetworking 21
Transmission Terminology
 Transmitter
 Receiver
 Medium
 Guided medium
• e.g. twisted pair, optical fiber
 Unguided medium
• e.g. air, water, vacuum
 Direct link
 No intermediate devices
 Point-to-point
• Only 2 devices share link
 Multi-point
• More than two devices share the link

Internetworking 22
Transmission Terminology
 Simplex
 One direction
• e.g. Television
 Half duplex
 Either direction, but only one way at a time
• e.g. police radio
 Full duplex
 Both directions at the same time
• e.g. telephone

Internetworking 23
Frequency, Spectrum and Bandwidth

 Time domain concepts

 Continuous signal
• Various in a smooth way over time
 Discrete signal
• Maintains a constant level then changes to another constant
level
 Periodic signal
• Pattern repeated over time
 Aperiodic signal
• Pattern not repeated over time

Internetworking 24
Continuous & Discrete Signals


Internetworking 25
Advanced Java Programming Periodic Signals

Internetworking 26
Advanced Java Programming Sine Wave

 Peak Amplitude (A)
 maximum strength of signal
 volts
 Frequency (f)
 Rate of change of signal
 Hertz (Hz) or cycles per second
 Period = time for one repetition (T)
 T = 1/f
 Phase (φ)
 Relative position in time

Internetworking 27
Varying Sine Waves


Internetworking 28
Advanced Java Programming Wavelength

 Distance occupied by one cycle
 Distance between two points of corresponding
phase in two consecutive cycles
 λ
 Assuming signal velocity v
 λ = vT
 λf = v
 c = 3*108 ms-1 (speed of light in free space)

Internetworking 29
Frequency Domain Concepts

 Signal usually made up of many frequencies

 Components are sine and/or cosine waves
 Can be shown (Fourier analysis) that any signal
is made up of component sine and/or cosine
waves
 Can plot frequency domain functions

Internetworking 30
Addition of Frequency Components


Internetworking 31
Frequency Domain

Internetworking 32
Spectrum & Bandwidth

 Spectrum
 range of frequencies contained in signal
 Absolute bandwidth
 width of spectrum
 Effective bandwidth
 Often just bandwidth
 Narrow band of frequencies containing most of the energy
 DC Component
 Component of zero frequency

Internetworking 33
Signal with DC Component


Internetworking 34
Data Rate and Bandwidth

 Any transmission system has a limited band of

frequencies
 This limits the data rate that can be carried

Internetworking 35
Analog and Digital Data Transmission

 Data
 Entities that convey meaning
 Signals
 Electric or electromagnetic representations of data
 Transmission
 Communication of data by propagation and processing of
signals

Internetworking 36
Advanced Java Programming Data
 Analog
 Continuous values within some interval
 e.g. sound, video
 Digital
 Discrete values
 e.g. text, integers

Internetworking 37
Acoustic Spectrum (Analog)


Internetworking 38
Advanced Java Programming Signals

 Means by which data are propagated
 Analog
 Continuously variable
 Various media
• wire, fiber optic, space
 Speech bandwidth 100Hz to 7kHz
 Telephone bandwidth 300Hz to 3400Hz
 Video bandwidth 4MHz
 Digital
 Use two DC components

Internetworking 39
Advanced Java ProgrammingData and Signals

 Usually use digital signals for digital data and
analog signals for analog data
 Can use analog signal to carry digital data
 Modem
 Can use digital signal to carry analog data
 Compact Disc audio

Internetworking 40
Analog Signals Carrying Analog and Digital Data


Internetworking 41
Digital Signals Carrying Analog and Digital Data


Internetworking 42
Analog Transmission
 Analog signal transmitted without regard to

content
 May be analog or digital data
 Attenuated over distance
 Use amplifiers to boost signal
 Also amplifies noise

Internetworking 43
Digital Transmission
 Concerned with content

 Integrity endangered by noise, attenuation etc.
 Repeaters used
 Repeater receives signal
 Extracts bit pattern
 Retransmits
 Attenuation is overcome
 Noise is not amplified

Internetworking 44
Advantages of Digital Transmission

 Digital technology
 Low cost LSI/VLSI technology
 Data integrity
 Longer distances over lower quality lines
 Capacity utilization
 High bandwidth links economical
 High degree of multiplexing easier with digital
techniques
 Security & Privacy
 Encryption
 Integration
 Can treat analog and digital data similarly

Internetworking 45
Transmission Impairments
 Signal received may differ from signal

transmitted
 Analog - degradation of signal quality
 Digital - bit errors
 Caused by
 Attenuation and attenuation distortion
 Delay distortion
 Noise

Internetworking 46
Advanced Java Programming Attenuation

 Signal strength falls off with distance
 Depends on medium
 Exponential in nature
 Issues
 Received signal strength:
• must be enough to be detected
• must be sufficiently higher than noise to be received without
error
 Attenuation is an increasing function of frequency
• Use equalization

Internetworking 47
Advanced Java Programming Delay Distortion

 Only in guided media
 Propagation velocity varies with
frequency

Internetworking 48
Advanced Java Programming Noise

 Additional signals inserted between
transmitter and receiver
 Thermal (Also called white noise)
 Due to thermal agitation of electrons
 Uniformly distributed
 Intermodulation
 Signals that are the sum and difference of original
frequencies sharing a medium
 Crosstalk
 A signal from one line is picked up by another
 Impulse
 Irregular pulses or spikes
 e.g. External electromagnetic interference
 Short duration
 High amplitude
Internetworking 49
Advanced Java ProgrammingChannel Capacity

 Data rate
 In bits per second
 Rate at which data can be communicated
 Bandwidth
 In cycles per second of Hertz
 Constrained by transmitter and medium
 Nyquist Bandwidth
 Establishes data rate and bandwidth for noise free channel
 Given M signal levels and B bandwidth, maximum data rate
C that can be achieved is
C = 2Blog2M
 Shannon’s Capacity
 Given signal to noise ratio SNR, maximum data rate
C = B log2(1 + SNR)
Transmission Media
Internetworking 51
Advanced Java Programming Overview

 Guided - wire
 Unguided - wireless
 Characteristics and quality determined by
medium and signal
 For guided, the medium is more important
 For unguided, the bandwidth produced by the
antenna is more important
 Key concerns are data rate and distance

Internetworking 52
Advanced Java Programming Design Factors

 Bandwidth
 Higher bandwidth gives higher data rate
 Transmission impairments
 Attenuation
 Interference
 Number of receivers
 In guided media
 More receivers (multi-point) introduce more attenuation

Internetworking 53
Electromagnetic Spectrum

Internetworking 54
Guided Transmission Media

 Twisted Pair
 Coaxial cable
 Optical fiber

Internetworking 55
Advanced Java Programming Twisted Pair

Internetworking 56
Twisted Pair - Applications

 Most common medium

 Telephone network
 Between house and local exchange (subscriber loop)
 Within buildings
 To private branch exchange (PBX)
 For local area networks (LAN)
 10Mbps or 100Mbps

Internetworking 57
Twisted Pair - Pros and Cons

 Cheap
 Easy to work with
 Low data rate
 Short range

Internetworking 58
Twisted Pair - Transmission Characteristics

 Analog
 Amplifiers every 5km to 6km
 Digital
 Use either analog or digital signals
 repeater every 2km or 3km
 Limited distance
 Limited bandwidth (1MHz)
 Limited data rate (100MHz)
 Susceptible to interference and noise

Internetworking 59
Unshielded and Shielded TP

 Unshielded Twisted Pair (UTP)

 Ordinary telephone wire
 Cheapest
 Easiest to install
 Suffers from external EM interference
 Shielded Twisted Pair (STP)
 Metal braid or sheathing that reduces interference
 More expensive
 Harder to handle (thick, heavy)

Internetworking 60
Advanced Java Programming UTP Categories

 Cat 3
 up to 16MHz
 Voice grade found in most offices
 Twist length of 7.5 cm to 10 cm
 Cat 4
 up to 20 MHz
 Cat 5
 up to 100MHz
 Commonly pre-installed in new office buildings
 Twist length 0.6 cm to 0.85 cm

Internetworking 61
Near End Crosstalk

 Coupling of signal from one pair to another

 Coupling takes place when transmit signal
entering the link couples back to receiving pair
 i.e. near transmitted signal is picked up by
near receiving pair

Internetworking 62
Advanced Java Programming Coaxial Cable

Internetworking 63
Coaxial Cable Applications

 Most versatile medium

 Television distribution
 Ariel to TV
 Cable TV
 Long distance telephone transmission
 Can carry 10,000 voice calls simultaneously
 Being replaced by fiber optic
 Short distance computer systems links
 Local area networks

Internetworking 64
Coaxial Cable - Transmission Characteristics

 Analog
 Amplifiers every few km
 Closer if higher frequency
 Up to 500MHz
 Digital
 Repeater every 1km
 Closer for higher data rates

Internetworking 65
Advanced Java Programming Optical Fiber

Internetworking 66
Optical Fiber - Benefits

 Greater capacity
 Data rates of hundreds of Gbps
 Smaller size & weight
 Lower attenuation
 Electromagnetic isolation
 Greater repeater spacing
 10s of km at least

Internetworking 67
Optical Fiber - Applications

 Long-haul trunks
 Metropolitan trunks
 Rural exchange trunks
 Subscriber loops
 LANs

Internetworking 68
Optical Fiber - Transmission Characteristics

 Act as wave guide for 1014 to 1015 Hz

 Portions of infrared and visible spectrum
 Light Emitting Diode (LED)
 Cheaper
 Wider operating temp range
 Last longer
 Injection Laser Diode (ILD)
 More efficient
 Greater data rate
 Wavelength Division Multiplexing

Internetworking 69
Optical Fiber Transmission Modes


Internetworking 70
Wireless Transmission
 Unguided media
 Transmission and reception via antenna
 Directional
 Focused beam
 Careful alignment required
 Omnidirectional
 Signal spreads in all directions
 Can be received by many antennae

Internetworking 71
Advanced Java Programming Frequencies
 2GHz to 40GHz
 Microwave
 Highly directional
 Point to point
 Satellite
 30MHz to 1GHz
 Omnidirectional
 Broadcast radio
 3 x 1011 to 2 x 1014
 Infrared
 Local

Internetworking 72
Terrestrial Microwave
 Parabolic dish
 Focused beam
 Line of sight
 Long haul telecommunications
 Higher frequencies give higher data rates

Internetworking 73
Satellite Microwave
 Satellite is relay station

 Satellite receives on one frequency, amplifies
or repeats signal and transmits on another
frequency
 Requires geo-stationary orbit
 Height of 35,784km
 Television
 Long distance telephone
 Private business networks

Internetworking 74
Advanced Java Programming Broadcast Radio

 Omnidirectional
 FM radio
 UHF and VHF television
 Line of sight
 Suffers from multipath interference
 Reflections

Internetworking 75
Advanced Java Programming Infrared

 Modulate noncoherent infrared light
 Line of sight (or reflection)
 Blocked by walls
 e.g. TV remote control, IRD port

Data Encoding
Internetworking 77
Encoding Techniques
 Data are not generally transmitted as they are

due to:
 Framing
 Error detection and correction
 Data are converted into transmittable signals
 There are two possibilities
 Analog communication
• Analog data, analog signal
• Telephone system
• Digital data, analog signal
• Computer to computer communication using telephone line
 Digital Communication
• Digital data, digital signal
• Analog data, digital signal

Internetworking 78
Digital Data, Digital Signal

 Digital signal
 Discrete, discontinuous voltage pulses
 Each pulse is a signal element
 Binary data encoded into signal elements

Internetworking 79
Advanced Java Programming Terminology

 Unipolar
 All signal elements have same sign
 Polar
 One logic state represented by positive voltage the other by
negative voltage
 Data rate
 Rate of data transmission in bits per second
 Bit duration or length of a bit
 Time taken for transmitter to emit the bit
 Modulation rate
 Rate at which the signal level changes
 Measured in baud = signal elements per second
 Mark and Space
 Binary 1 and Binary 0 respectively
Internetworking 80
Interpreting Signals
 Need to know
 Timing of bits - when they start and end
 Signal levels
 Factors affecting successful interpreting of
signals
 Signal to noise ratio
 Data rate
 Bandwidth

Internetworking 81
Comparison of Encoding Schemes

 Signal Spectrum
• Lack of high frequencies reduces required bandwidth
• Lack of dc component allows ac coupling via transformer,
providing isolation
• Concentrate power in the middle of the bandwidth
 Clocking
• Synchronizing transmitter and receiver
• External clock
• Sync mechanism based on signal
 Error detection
• Can be built in to signal encoding
 Signal interference and noise immunity
• Some codes are better than others
 Cost and complexity
• Higher signal rate (& thus data rate) lead to higher costs
• Some codes require signal rate greater than data rate
Internetworking 82
Encoding Schemes
 Non-Return to Zero
 Non-Return to Zero-Level (NRZ-L)
 Non-Return to Zero Inverted (NRZ-I)
 Multilevel Binary
 Bipolar -AMI
 Pseudoternary
 Biphase
 Manchester
 Differential Manchester
 Scrambling
 B8ZS
 HDB3

Internetworking 83
Non-Return to Zero Level (NRZ-L)

 Two different voltages for 0 and 1 bits

 Voltage constant during bit interval
 no transition i.e. no return to zero voltage
 e.g. Absence of voltage for zero, constant
positive voltage for one
 More often, negative voltage for one value and
positive for the other
 Example
0 1 0 0 1 1 0 0 0 1 1

Internetworking 84
Differential Encoding
 Data represented by changes rather than

levels
 More reliable detection of transition rather
than level
 In complex transmission layouts it is easy to
lose sense of polarity

Internetworking 85
Non-Return to Zero Inverted(NRZ-I)

 Non-return to zero inverted on ones

 Constant voltage pulse for duration of bit
 Data encoded as presence or absence of signal
transition at beginning of bit time
 Transition (low to high or high to low) denotes
a binary 1
 No transition denotes binary 0
 An example of differential encoding
0 1 0 0 1 1 0 0 0 1 1

Internetworking 86
NRZ pros and cons

 Pros
 Easy to engineer
 Make good use of bandwidth
 Cons
 dc component
 Lack of synchronization capability
 Lack of error detection/correction facility
 Used for magnetic recording
 Not often used for signal transmission

Internetworking 87
Advanced Java ProgrammingMultilevel Binary

 Use more than two levels
 Bipolar-AMI
 zero represented by no line signal
 one represented by positive or negative pulse alternatively
 one pulses alternate in polarity
 Pros
 No loss of sync if a long string of ones (zeros still a problem)
 No net dc component
 Lower bandwidth
 Easy error detection
0 1 0 0 1 1 0 0 0 1 1

Internetworking 88
Advanced Java Programming Pseudoternary

 One represented by absence of line signal
 Zero represented by alternating positive and
negative
 No advantage or disadvantage over bipolar-
AMI
0 1 0 0 1 1 0 0 0 1 1

Internetworking 89
Bipolar-AMI and Pseudoternary


Internetworking 90
Trade Off for Multilevel Binary

 Not as efficient as NRZ

 Each signal element only represents one bit
 In a 3 level system could represent log23 = 1.58 bits
 Receiver must distinguish between three levels
(+A, -A, 0)
 Requires approx. 3dB more signal power for same
probability of bit error

Internetworking 91
Advanced Java Programming Biphase

 Manchester
 Transition in middle of each bit period
 Transition serves as clock and data
 Low to high represents one
 High to low represents zero
 Used by IEEE 802.3
0 1 0 0 1 1 0 0 0 1 1

Internetworking 92
Advanced Java Programming Biphase

 Differential Manchester
 Midbit transition is clocking only
 Transition at start of a bit period represents zero
 No transition at start of a bit period represents one
 Note: this is a differential encoding scheme
 Used by IEEE 802.5
0 1 0 0 1 1 0 0 0 1 1

Internetworking 93
Biphase Pros and Cons

 Pros
 Synchronization on mid bit transition (self clocking)
 No dc component
 Error detection
• Absence of expected transition
 Con
 At least one transition per bit time and possibly two
 Maximum modulation rate is twice NRZ
 Requires more bandwidth

Internetworking 94
Advanced Java Programming Scrambling

 Use scrambling to replace sequences that
would produce constant voltage
 Filling sequence
 Must produce enough transitions to sync
 Must be recognized by receiver and replace with original
 Same length as original
 No dc component
 No long sequences of zero level line signal
 No reduction in data rate
 Error detection capability

Internetworking 95
Advanced Java Programming B8ZS

 Bipolar With 8 Zeros Substitution
 Based on bipolar-AMI
 If octet of all zeros and last voltage pulse
preceding was positive encode as 000+-0-+
 If octet of all zeros and last voltage pulse
preceding was negative encode as 000-+0+-
 Causes two violations of AMI code
 Unlikely to occur as a result of noise
 Receiver detects and interprets as octet of all
zeros

Internetworking 96
Advanced Java Programming HDB3

 High Density Bipolar 3 Zeros
 Based on bipolar-AMI
 String of four zeros replaced with one or two
pulses
Number of Bipolar pulses(ones) since

last substitution
Polarity of Odd Even
preceding pulse
- 000- +00+
+ 000+ -000-

Internetworking 97
Advanced Java Programming B8ZS and HDB3

Internetworking 98
Bandwidth Comparison

Internetworking 99
Advanced Java Programming Modulation Rate

Internetworking 100
Advanced Java Programming Modulation Rate
Minimum 10101010… Maximum

NRZ-L 0 (all 0s or 1s) 1.0 1.0
NRZ-I 0 (all 0s) 0.5 1.0 (all 1s)
Bipolar AMI 0 (all 0s) 1.0 1.0
Pseudoternary 0 (all 1s) 1.0 1.0
Manchester 1.0 (10101…) 1.0 2.0 (all 0s or 1s)
Differential 1.0 (all 1s) 1.5 2.0 (all 0s)
manchester

Data Link Layer
Internetworking 102
Data Link Layer Design Issues

 Services provided to the Network Layer

 Unacknowledged connectionless service
 Acknowledged connectionless service
 Acknowledged connection-oriented service
 Accessing of underlying medium (MAC sub-layer)
 Data Link Control(Data Link Control Sub-layer)
 Framing
 Identify beginning and ending of a frame
 Error Detection & Correction
 Identifying transmission errors and if possible correction
 Error Control
 What to do if frames/acknowledgements are damaged/lost?
 Flow Control
 Transmitter and Receiver must be synchronized
Medium Access
Control Sub-layer
Internetworking 104
Network Layer
LLC sub-layer
Data Link Layer
MAC sub-layer
Physical Layer

Internetworking 105

 Channel
 Unicast—single user
 Broadcast—sometimes called multi-access/random
access channels
 Key issue in a broadcast network
 determine who gets to use the channel when there is
competition for it
 the protocols used to determine who goes next on a
multiaccess channel belong to a sublayer of the data
link layer called the MAC (Medium Access
Control) sublayer.

Internetworking 106
Channel Allocation Problem

 Two schemes exist to solve the problem of allocation

for a single broadcast channel among competitive
users:
 static channel allocation
 dynamic channel allocation
 The traditional way of allocating a single channel

 Frequency Division Multiplexing Access(FDMA)
• Each station is allocated its own frequency
• Uneven usage of bandwidth if number of senders is large and
continuously varying
 Time Division Multiplexing Access (TDMA)
• Each station is allocated its own time slot for transmission
 Statistical Time Division Multiplexing Access (STDMA)
Internetworking 107
Problems of Static Allocation

 FDMA, TDMA are inherently inefficient

 C=channel capacity in bps
 λ=frame arrival rate in frames/sec
 1/µ=mean frame size (random/Poison distribution)
1
 Mean time delay from queueing theory
T =
µC −λ
 For FDMA
1 N
 TFDMA = = = NT
µ ( C / N ) − (λ / N ) µC −λ
 For TDMA
1 N
 TTDMA = = = NT
µ ( C / N ) − (λ / N ) µC −λ
 The mean delay for FDMA or TDMA is N times worse
Internetworking 108
Dynamic Channel Allocation

 Five key assumptions:

1. station model
 N number of independent stations.
 Frame generation rate is proportional to ∆t is ∆tλ
 Blocks after a frame generation until it is transmitted
2. single channel
 Only one channel is available for all stations
3. collision assumption
 Collision can occur. Every station can detect collision
4a. continuous time
4b. slotted time
5a. carrier sense
 Sense the carrier before transmission
5b. no carrier sense
 transmit without sensing the carrier
Internetworking 109
Advanced Java Programming ALOHA

 Pure ALOHA
 Slotted ALOHA

Internetworking 110
Advanced Java Programming Pure ALOHA

 A station transmits a frame whenever it has
one
 If collision occurs,
 detect it
 wait a random amount of time and retransmit

Internetworking 111
Efficiency of Pure ALOHA

 Frame time—time required to transmit a single

frame
 S=mean number of frame generated per frame
time (Poision)
 G=mean number of frames (old and new combined)
transmitted
 P0=probability that a frame does not suffer a
collision
 Throughput S=GP0
 Probability the k frames arekgenerated
−G
during a
G edistribution:
given frame time by Poisson
Pk =
K!
Internetworking 112
Efficiency of Pure ALOHA

P0 = e − G
 Throughpu S = Ge −2G
1
 tMaximum S max = with G = 0.5
throughput
2e
 Out of 100 frames, maximum of 18 frames reach
their destination without collision
Internetworking 113
Advanced Java Programming Slotted ALOHA

 Time is divided into discrete intervals
 Transmission of a frame is only allowed at
the beginning of a slot
 If a frame is generated in the middle of a
slot, the station must wait for the next slot
 Throughpu S = Ge − G
t 1
 Maximum S max = with G = 1
throughput
e

Internetworking 114
Throughput of ALOHA

Internetworking 115
Throughput of ALOHA
 Probability that a frame will avoid collision is e-G

 Probability that a frame will suffer a collision is (1-e-
G
)
 Probability of a transmission requires exactly k
attempts is
 Pk= e-G (1-e-G)k-1
α α

E = kP =
Expected
∑
number
k =1
of −G
ke (1 − e )
transmission
k ∑
−G k −1
k =1
= e G

Internetworking 116
Advanced Java Programming CSMA Protocols

• Protocols in which stations listen for a carrier and act
accordingly are called carrier sense protocols
• Persistent CSMA
• If the channel is busy, station waits until it becomes idle (it
continually senses the carrier for the purpose of seizing it)
• Non-Persistent CSMA
• If the channel is busy, station waits random amount of time
and repeats the algorithm
• 1-persistent CSMA
• If the channel is idle, the station transmits with a
probability of 1
• If collision occurs, it waits a random amount of time, and
starts all over again
• p-persistent
• applies to slotted channels, when the station becomes
ready to send, it senses the channel, if it is idle, it
transmits with a probability p
Internetworking 117
Advanced Java Programming CSMA/CD Protocol
• If collision occurs, stop transmission immediately

Internetworking 118
Advanced Java Programming Collision

Internetworking 119
Advanced Java Programming Collision Detection

Internetworking 120
Advanced Java Programming CSMA/CD Protocol

Internetworking 121
Performance Comparison

Internetworking 122
Collision Free Protocols

 Collision adversely affect system performance,

when cable is long and frames short and number of
stations is large
 Collision free protocols is the solution
 Bit-Map Protocol
 Binary Countdown
 Basic Assumptions
 N number of stations
 Each station has a unique address from 0 to N-1
 Which station gets the channel next?

 Reservation protocols.
Internetworking 123
Advanced Java Programming Bit-Map Protocol

 Basic Steps
 Contention period
 Frame transmission
 Contention slot has exactly N slots one for each
station
 Station j is allowed to transmit (either 1 or 0) during
slot j
 No other station is allowed to transmit during that
slot
 Station j sends a 1 if it has frame to transmit, 0
otherwise
 After N time slots, each station has complete
knowledge of which stations wish to transmit
© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:24
Internetworking 124
Advanced Java Programming Bit-Map Protocol
 If a station is ready, after its bit slot, it must wait for

next bit map

Internetworking 125
Performance of Bit-Map Protocol

 Assumptions
 Duration of each contention bit slot is unity
 Duration of a data frame is d time units
 High numbered stations must wait on the average
0.5N slots
 Low numbered stations must wait on the average
1.5N slots
 mean for all stations is N slots
 For low load, the overhead per frame is N bits, and
the amount of data is d bits, for an efficiency of
d/(N+d).
 For high load, the overhead per frame is 1 bit, and
the amount of data is Nd bits, for an efficiency of
d/(d+1)
Internetworking 126
Binary Countdown Protocol

 1 bit overhead per station in Bit-Map protocol

 Solution
 Use binary station address
 Basic idea
 A station with highest address will be allowed to transmit
 All addresses are assumed to be same length
 Stations broadcast their address bits in each slot, stating
with high order bit
 The bits are then BOOLEAN ORed
 If a station sending a 0, gets 1 (this means that there is at
least one station with higher address), it gives up

Internetworking 127
Binary Countdown Protocol


Internetworking 128
Efficiency of Binary Countdown Protocol

• The channel efficiency of this method, under low load is

d/(d+ lnN).
• The channel efficiency of this method, under high load is
Nd/(Nd+ lnN).

Internetworking 129
Problems of Binary Countdown Protocol

 Starvation may occur

 Solution proposed by Mok and Ward
 Station having virtual address will be permuted circularly
 Example
 C, H, D, A, G, B, E, F have priorities 7, 6, 5, 4, 3, 2, 1, 0
 If D transmits, it will removed from this order and put at
the end giving a priority order C, H, D, A, G, B, E, F, D

Internetworking 130
Performance Comparison
 Two basic protocols

 Contention
 Contention free
 Two performance measures
 Delay
 Channel efficiency
 Contention protocols are preferable at low load
 Contention free protocols are preferable at high load
 Solution is Limited Contention Protocol

Internetworking 131
Symmetric Protocol
 k number of stations
 Each has a probability p of transmitting during each
slot
 Probability that some station will successfully
acquire the channel during a given slot is A=kp(1-
p)k-1
 pmax= 1/k
 Amax=(1-1/k)k-1

Internetworking 132
Advanced Java ProgrammingSymmetric Protocol

Internetworking 133
Limited Contention Protocol

 Probability of channel acquisition can be increased

by reducing competition
 Basic idea
 Divide the station into groups
 At slot 0, members of group 0 is permitted to transmit
 If one of them succeeds, it transmits
 If slot 0 is empty or there is collision, members of group 1
contend for slot 1, etc.
 By making appropriate division, amount of
contention can be reduced.
 Choice
 Each group has one member—contention free protocols
 All station are in single group—ALOHA/Slotted ALOHA

Internetworking 134
Adaptive Tree Walk Protocol

 Number of members in a group a dynamically

changed
 Example
 All station are allowed to transmit at slot 0
 If there is no collision, next slot 1 will be used by all stations
again
 Otherwise, stations will be broken into two groups each
having half number of stations.
 Slot 1 will be used by members of first half etc.

Internetworking 135
Advanced Java Programming Example

Internetworking 136
Advanced Java Programming Improvement
• Each node at level I has a fraction 2-i of the stations below it

• if the q ready stations are uniformly distributed, the expected number of them
below a specific node at level i is just 2-iq
• the optimum level to begin searching the tree is at i=log2q
Internetworking 137
Further Improvement
 G and H want to transmit

 At slot 0, collision will occur, slot 1 will be idle
 It is pointless to probe node 3 (because collision will
be the obvious result)

Data Link Control
Sub-layer
Internetworking 139
Advanced Java Programming Framing

 Breaks the message into number of segments called
frames
 This is done to
 reduce total transmission time
S A B D S A B D
1
Transmission
delay message 2
Total 3 1
Delay 2
TD 3 1
Total
message 2
Delay TD
3
Propagation delay
message
Time Time
 reduce amount of retransmission due to error

Internetworking 140
Optimal Number of frames?

S A B D x = number of bits in the message

h = header length
k = number of hops between source and destination
R = data rate
1 t p = propagation delay
tm p = number of packets to be deter mined
2 x 
p + h 
1
tf = 
p 
3
R
x 
 + h
2 TD t =
p 
1 f R
3
x 
tp 2 p + h 
p
tm =  
tf 3 R
T x  x 
p + h   + h 
p  + (k − 1)  p
TD =   + kt
p
R R
dTD x(k − 1)
=0 ⇒ p=
dp h
Internetworking 141
Expected amount of transmission?

 x bit message
 b=bit error
 rate (1 − b)
Probability that the message will not be in error is
x
 Probability that the message will be in 1 − (1 − b) x
error is
 Probability that the message requires exactly i
transmission for successful transmission [i.e. (i-1)
unsuccessful transmission followed[1 (1 − b) x ]( i −1) (1 − b) x
by−successful
transmission] is
 Expected number of
transmission i =α
N rx = ∑ i[1 − (1 − b) ]
x ( i −1) x
(1 − b)
i =1
1
N rx =
(1 − b) x x
 Total number of bits transmitted without
framing (1 − b) x
Internetworking 142
Expected amount of transmission?

 p=number of frames
1
 Expected number of N rp =
transmission (1 − b ) x/ p
 Number of bits transmitted with

framing ( x / p ) x
p =
(1 − b) x/ p
(1 − b) x/ p
number of bits transmitted without fram ing (1 − b) x / p

=ω =
number of bits transmitted with fram ing (1 − b) x
1
= x −( x / p )
(1 − b )
 x=10,000 bits
 p=10
 b=0.001 (1 out of 1000)
 ω= 8139

Internetworking 143
Advanced Java Programming Framing

 four methods for framing are:
 character count
 starting and ending characters
 starting and ending flags
 physical layer coding violations

Internetworking 144
Advanced Java Programming Character Count

 Assumes character oriented data transmission
 This method uses a field in the header to specify the
number of characters in the frame
 When data link layer at the destination sees this, it
knows how many characters follow

Internetworking 145
Starting and ending characters

 Gets around the problem of resynchronization

 Each frame starts with special ASCII character sequence DLE
STX and ends with the sequence DLE ETX
 DLE is Data Link Escape, STX is Start of TeXt and ETX is End of
TeXt
 If destination ever loses track of frame boundaries, all it has to
do is look for DLE STX or DLE ETX to figure out where it is
DLE STX A B D DLE ETX DLE STX D B DLE ETX

Internetworking 146
Starting and ending characters

 Problem occurs when data contains DLE STX or DLE ETX

DLE STX A DLE D DLE ETX Data sent by network layer
 Solution
 Sender’s data link layer inserts an ASCII DLE character just before
an “accidental” DLE character in the data
 This technique is called character stuffing
Data after being character stuffed

DLE STX A DLE DLE D DLE ETX
by data link layer
 Data link layer at receiving end removes the DLE character

 Thus framing DLE STX or DLE ETX can be distinguished by absence
or presence of a single DLE as DLEs in the data are always doubled
Data passed to network layer on the

DLE STX A DLE D DLE ETX
receiving node

Internetworking 147
Starting and ending flags

Bit Stuffing

Internetworking 148
Advanced Java Programming Error Detection

 The data link layer breaks the bit stream up into
discrete frames and then computes the checksum for
each frame
 when a frame arrives at the destination the checksum is
recomputed, and if it is different from the one contained
in the frame, the data link layer knows that an error has
occurred

Internetworking 149
Advanced Java Programming Error Detection
 Additional bits added by transmitter for error detection

Internetworking 150
Error Detection Techniques

 Parity
 Value of parity bit is such that character has even (even
parity) or odd (odd parity) number of ones
 Even number of bit errors goes undetected

Internetworking 151
Two-dimensional parity
Detectable
Double-bit-
error
1 0 1 0 1 1
Undetec
1 1 0 1 1 0 ed
error
0 1 1 1 0 1
0 0 1 0 1 0
Detected
error

Internetworking 152
Cyclic Redundancy Check(CRC)

 For a block of d bits transmitter generates r bit sequence

 Transmit d+r bits which is exactly divisible by some
predetermined number G
 Receiver divides frame by G
 If no remainder, no error

Internetworking 153
Cyclic Redundancy Check (CRC)

 D=d-bit data
 G=(r+1)-bit predetermined divisor
 F=r-bit FCS to be determined
 T=(d+r)-bit message to be transmitted
r
D.2 R
=Q+
G G
T = D.2 + R r
T D.2 + R R R R+R
r
= =Q+ + =Q+ =Q
G G G G G
Internetworking 154

D=1010001101
G=110101
R=? G 1 1 0 1 0 1|1 0 1 0 0 0 1 1 0 1 0 0 0 0 0
110101
111011
110101
1110 10
1101 01
11 11 10
11 01 01
101100
T=101000110101110 110101
110010
110101
0 1 1 1 0R
Internetworking 155

D=101110
G=1001
R=?

Internetworking 156
Error Correcting codes

 m-bit message
 k redundant bits are added
 T=(m+k)-bit message to be transmitted
 No of possible code words with (m+k)-bit is 2m+k
 Out of 2m+k possible code words, only 2m code words are
valid
 Remaining code words are invalid
 Example
 M=2, K=2
 No. of possible codewords is 24 =16
 No. of valid code words is 4
• 0000, 0011, 1100, 1111
 No. of invalid code words is 12
0000 0100

Internetworking 157
Advanced Java Programming Error Correction
•Hamming codes can correct single errors

•by arranging them into matrix we can correct burst
errors
Internetworking 158

Internetworking 159
 The polynomial code (cyclic redundancy code or CRC

code) are used for error detecting
 when the polynomial code method is employed, the

sender and receiver must agree upon a generator
polynomial, G(x), in advance.
 Both high and low bits of the generator must be 1
 to compute the checksum for some frame with m bits,

corresponding to the polynomial M(x), the frame must
be longer than the generator polynomial

Internetworking 160

 The algorithm for computing the checksum is as follows:
 let r be the degree of G(x). Append r zero bits to the
low-order end of the frame, so it now contains m+r
bits and corresponds to the polynomial xrM(x)
 divide the bit string corresponding to G(x) into the
bit string corresponding to xrM(x) using modulo 2
division
 subtract the remainder (which is always r or fewer
bits) from the bit string corresponding to xrM(x)
using modulo 2 subtraction. The result is the
checksummed frame to be transmitted. Call its
polynomial T(x).

Internetworking 161
Advanced Java Programming Flow Control

 Receiver may be slower than transmitter
 Transmitter must not transmit frames at a rate faster
than the receiver can receive
 Ensuring the sending entity does not overwhelm the
receiving entity
 Preventing buffer overflow
 Transmission time
 Time taken to emit all bits into medium
 Propagation time
 Time for a bit to traverse the link
propagation time
a=
transmission time
 If transmission time is normalized to 1, propagation time is `a`

Internetworking 162
Model of Frame Transmission


Internetworking 163
Advanced Java Programming Stop and Wait

 Source transmits frame
 Destination receives frame and replies with
acknowledgement
 Source waits for ACK before sending next
frame
 Destination can stop flow by not sending ACK
 Works well for a few large frames

Internetworking 164
Advanced Java Programming Fragmentation

 Large block of data may be split into small
frames
 Limited buffer size
 Errors detected sooner (when whole frame received)
 On error, retransmission of smaller frames is needed
 Prevents one station occupying medium for long periods
 Stop and wait becomes inadequate

Internetworking 165
Stop and Wait Link Utilization

TD=tframe+tprop+tproc+tack+tprop+tpr
1
tframe
a oc
tprop
TD≈tframe+2tprop
tproc transmission time
η=
tack ack 1 propagation time
t frame
=
2
t frame + 2t prop
1
=
t prop
1+ 2
ack 2 t frame
1 t prop
= where a =
1 + 2a t frame
Internetworking 166


Internetworking 167

 tprop=271 ms
 1 kb frame
 1 Mbps data rate
 Frame transmission time tframe=1 ms
 a=271
 η= 1/(1+2x271)=0.00184

Internetworking 168
Sliding Window Flow Control

 Allow multiple frames to be in transit at a time

 Receiver has buffer W long
 Transmitter can send up to W frames without
ACK
 Each frame is numbered
 ACK includes number of next frame expected
 Sequence number bounded by size of field (k)
 Frames are numbered modulo 2k

Internetworking 169
Sliding Window Diagram


Internetworking 170
Example Sliding Window


Internetworking 171
Sliding Window Enhancements

 Receiver can acknowledge frames without

permitting further transmission (Receive Not
Ready)
 Must send a normal acknowledge to resume
 If duplex, use piggybacking
 If no data to send, use acknowledgement frame
 If data but no acknowledgement to send, send last
acknowledgement number again, or have ACK valid flag
(TCP)

Internetworking 172
Advanced Java Programming Error Control

 Detection and correction of errors
 Lost frames
 Damaged frames
 Automatic repeat request
 Error detection
 Positive acknowledgment
 Retransmission after timeout
 Negative acknowledgement and retransmission

Internetworking 173
Automatic Repeat Request (ARQ)

 Stop and wait

 Go back N
 Selective reject (selective retransmission)

Internetworking 174
Advanced Java Programming Stop and Wait

 Source transmits single frame
 Wait for ACK
 If received frame damaged, discard it
 Transmitter has timeout
 If no ACK within timeout, retransmit
 If ACK damaged,transmitter will not recognize
it
 Transmitter will retransmit
 Receive gets two copies of frame
 Use ACK0 and ACK1

Internetworking 175
Stop and Wait -Diagram


Internetworking 176
Stop and Wait - Pros and Cons

 Simple
 Inefficient

Internetworking 177
Advanced Java Programming Go Back N

 Based on sliding window
 If no error, ACK as usual with next frame
expected
 Use window to control number of outstanding
frames
 If error, reply with rejection
 Discard that frame and all future frames until error frame
received correctly
 Transmitter must go back and retransmit that frame and all
subsequent frames

Internetworking 178
Go Back N - Damaged Frame

 Receiver detects error in frame i

 Receiver sends rejection-i
 Transmitter gets rejection-i
 Transmitter retransmits frame i and all
subsequent

Internetworking 179
Go Back N - Lost Frame (1)

 Frame i lost
 Transmitter sends i+1
 Receiver gets frame i+1 out of sequence
 Receiver send reject i
 Transmitter goes back to frame i and
retransmits

Internetworking 180
Go Back N - Lost Frame (2)

 Frame i lost and no additional frame sent

 Receiver gets nothing and returns neither
acknowledgement nor rejection
 Transmitter times out and sends
acknowledgement frame with P bit set to 1
 Receiver interprets this as command which it
acknowledges with the number of the next
frame it expects (frame i )
 Transmitter then retransmits frame i

Internetworking 181
Go Back N – Damaged Acknowledgement

 Receiver gets frame i and send

acknowledgement (i+1) which is lost
 Acknowledgements are cumulative, so next
acknowledgement (i+n) may arrive before
transmitter times out on frame i
 If transmitter times out, it sends
acknowledgement with P bit set as before
 This can be repeated a number of times before
a reset procedure is initiated

Internetworking 182
Go Back N - Damaged Rejection

 As for lost frame (2)

Internetworking 183
Go Back N - Diagram

Internetworking 184
Maximum window size?

 Consider a piggybacking acknowledgement

scheme is used
 3-bit sequence number
 Transmitter sends frame 0 and gets RR1
 Transmitter then sends 1, 2, 3, 4, 5, 6, 7, 0
 Transmitter gets RR1
 What does it mean?
 All eight frames are received properly
 All are lost and receiver is repeating its previous RR1
 For n bit sequence number maximum window

size can be 2n-1

Internetworking 185
Selective Reject/Retransmission
 Only rejected frames are retransmitted

 Subsequent frames are accepted by the
receiver and buffered
 Minimizes retransmission
 Receiver must maintain large enough buffer
 More complex login in transmitter

Internetworking 186
Selective Reject -Diagram


Internetworking 187
Maximum window size?

 Transmitter sends frame 0 through 6

 All 7 frames are received properly.
 Receiver advances its window to accept frames
7 through 5 and sends RR7
 RR7 is lost in transit.
 Transmitter times out and retransmit frame 0
 Receiver accepts this frame as new frame
 For n bit sequence number maximum window

size can be 2n-1

Internetworking 188
Advanced Java Programming Performance?

Internetworking 189

Internetworking 190
1 N ≥ 1 + 2a

U =  N
1 + 2a N < 1 + 2a

Internetworking 191

Tf
 Channel Utilization U =
Tt
 Where Tf=time to transmit a single frame
 Tt=total time that line is engaged in the
transmission of a single frame
 For error free operation using Stop-and-Wait
protocol Tf
U=
T f + 2T p
 In the presence of error equation must be modified
T
as U=
f
N r (T f + 2T p )
 Where Nr=expected number of transmission for a

frame

Internetworking 192

 p=probability that a single frame is in error
 Assume ACKs and NAKs are never in error
 To transmit a frame successfully, it requires
exactly i attempts
 That means i-1 times unsuccessful (with error)
transmission followed by 1 successful (without
error) transmission
 Probability that a frame requires exactly i
transmission is pi-1(1-p) α
 Nr=E[transmission]= ∑ (1 − p)
ip i −1
i =1
1
=
1− p
Internetworking 193

1− p
 Stop-and-Wait ARQ U =
1 − 2a
1 − p N ≥ 1 + 2a
 Selective Reject ARQ U = 
 N (1 − p)
 1 + 2a N < 1 + 2a

Internetworking 194
Advanced Java Programming Go-back-N?

 Nr=E[number of transmitted frame to transmit one frame
successfully]
 f(i)=total number of transmission if original frame
must be transmitted i times
 K=total number of frames retransmitted (including
the original frame) for each
α
error
N r = ∑ [(1 − K ) + Ki ] p i −1 (1 − p )
 f(i)=1+(i-1)K i =1
α α
 =(1-K)+Ki = (1 − K )∑ p (1 − p ) + K ∑ ip i −1 (1 − p )
i −1
i =1 i =1
K
= (1 − K ) +
1− p
1 − p + Kp
=
1− p

Internetworking 195
Advanced Java Programming Go-back-N?

 1− p N ≥ 1 + 2 a , K = 1 + 2 a
1 + 2ap
U = 
 N (1 − p)
N < 1 + 2a, K = N
(1 + 2a )(1 − p + Np )

Internetworking 196

Internetworking 197
High Level Data Link Control

 Widely used data link control protocol

 Basic Characteristics
 Three types of stations
 Two link configurations
 Three data transfer mode

Internetworking 198
HDLC Station Types

 Primary station
 Controls operation of link
 Maintains separate logical link to each secondary station
 Frames issued are called commands
 Secondary station
 Operates under control of primary station
 Frames issued called responses
 Combined station
 May issue commands and responses

Internetworking 199
HDLC Link Configurations

 Unbalanced
 One primary and one or more secondary stations
 Supports full duplex and half duplex
 Balanced
 Two combined stations
 Supports full duplex and half duplex

Internetworking 200
HDLC Transfer Modes

 Normal Response Mode (NRM)

• Unbalanced configuration
• Host computer as primary, Terminals as secondary
• Primary initiates transfer to secondary
• Secondary may only transmit data in response to command
from primary
• Used on multi-drop lines
 Asynchronous Balanced Mode (ABM)
• Balanced configuration
• Either station may initiate transmission without receiving
permission
• Most widely used
• No polling overhead
 Asynchronous Response Mode (ARM)
• Unbalanced configuration
• Secondary may initiate transmission without permission form
primary
• Primary responsible for line
• rarely used

Internetworking 201
Advanced Java Programming Frame Structure

 Uses synchronous transmission
 All transmissions in frames
 Single frame format for all data and control
exchanges

Internetworking 202
Advanced Java Programming Flag Fields

 Delimit frame at both ends
 01111110
 May close one frame and open another
 Receiver hunts for flag sequence to
synchronize
 Bit stuffing used to avoid confusion with data
containing 01111110
 0 inserted after every sequence of five 1s
 If receiver detects five 1s it checks next bit
 If 0, it is deleted
 If 1 and seventh bit is 0, accept as flag
 If sixth and seventh bits 1, sender is indicating abort

Internetworking 203
Advanced Java Programming Bit Stuffing
 Example with
possible errors

Internetworking 204
Advanced Java Programming Address Field

 Identifies secondary station that will send or
will receive frame
 Usually 8 bits long
 May be extended to multiples of 7 bits
 LSB of each octet indicates that it is the last octet (1) or not
(0)
 All ones (11111111) is broadcast

Internetworking 205
Advanced Java Programming Control Field

 Different for different frame type
 Information - data to be transmitted to user (next layer up)
• Flow and error control piggybacked on information frames
 Supervisory - ARQ when piggyback not used
 Unnumbered - supplementary link control
 First one or two bits of control filed identify
frame type
 Remaining bits explained later

Internetworking 206
Control Field Diagram


Internetworking 207
Supervisory frames

Internetworking 208
Unnumbered frames

Internetworking 209
Advanced Java Programming Poll/Final Bit

 Use depends on context
 Command frame
 P bit
 1 to solicit (poll) response from peer
 Response frame
 F bit
 1 indicates response to soliciting command

Internetworking 210
Information Field
 Only in information and some unnumbered

frames
 Must contain integral number of octets
 Variable length

Internetworking 211
Frame Check Sequence Field

 FCS
 Error detection
 16 bit CRC
 Optional 32 bit CRC

Internetworking 212
Advanced Java Programming HDLC Operation

 Exchange of information, supervisory and
unnumbered frames
 Three phases
 Initialization
 Data transfer
 Disconnect

Internetworking 213
Examples of Operation (1)


Internetworking 214
Examples of Operation (2)


Internetworking 215
Other DLC Protocols (LAPB,LAPD)

 Link Access Procedure, Balanced (LAPB)

 Part of X.25 (ITU-T)
 Subset of HDLC - ABM
 Point to point link between system and packet switching
network node
 Link Access Procedure, D-Channel
 ISDN (ITU-D)
 ABM
 Always 7-bit sequence numbers (no 3-bit)
 16 bit address field contains two sub-addresses
• One for device and one for user (next layer up)

Internetworking 216
Other DLC Protocols (LLC)

 Logical Link Control (LLC)

 IEEE 802
 Different frame format
 Link control split between medium access layer (MAC) and
LLC (on top of MAC)
 No primary and secondary - all stations are peers
 Two addresses needed
• Sender and receiver
 Error detection at MAC layer
• 32 bit CRC
 Destination and source access points (DSAP, SSAP)

Internetworking 217
Other DLC Protocols (Frame Relay)

 Streamlined capability over high speed packet

witched networks
 Used in place of X.25
 Uses Link Access Procedure for Frame-Mode
Bearer Services (LAPF)
 Two protocols
 Control - similar to HDLC
 Core - subset of control
 ABM
 7-bit sequence numbers
 16 bit CRC
 2, 3 or 4 octet address field
 Data link connection identifier (DLCI)
 Identifies logical connection
 More on frame relay later
Internetworking 218
Other DLC Protocols (ATM)

 Asynchronous Transfer Mode

 Streamlined capability across high speed
networks
 Not HDLC based
 Frame format called “cell”
 Fixed 53 octet (424 bit)
 Details later

Internetworking 219
Required Reading
 Stallings chapter 7
 Web sites on HDLC, frame relay, Ethernet and
ATM

Local Area
Networks(LANs)
IEEE802.3
(Ethernet) LAN
Internetworking 222
Advanced Java Programming Ethernet Cabling

Internetworking 223
Advanced Java Programming Ethernet Topology

Internetworking 224
Advanced Java Programming Encoding

Internetworking 225
Ethernet Frame format


Internetworking 226
Advanced Java Programming Collision detection

Internetworking 227
Advanced Java Programming Ethernet Efficiency

Internetworking 228
Switched Ethernet LAN


IEEE802.4
(Token Bus) LAN
Internetworking 230
Advanced Java Programming Token Bus
IEEE Standard 802.4

Internetworking 231
Token Bus Frame Format


Internetworking 232
Token Bus Control Frames


IEEE802.5
(Token Ring) LAN
Internetworking 234
IEEE Standard 802.5: Token Ring

Internetworking 235

Internetworking 236

Internetworking 238

Internetworking 239

Internetworking 240
Token Ring Frame Format


Internetworking 241
Token Ring Control Frames


Wide Area
Networks(WANs)
IEEE802.6
(DQDB) LAN
Internetworking 244

Network Layer
Internetworking 246
Network Layer Design Issues

 The network layer is concerned with getting packets

from all the way to the destination.
 The network layer is the lowest layer that deals with
end-to-end transmission
 the network layer provides services to the transport
layer at the network layer/transport layer interface.
 The network layer services have been designed with
the following goals in mind:
 The services should be independent of the subnet
technology
 the transport layer should be shielded from the
number, the type and the topology of the subnets
present
 the network addresses made available to the
transport layer should use a uniform numbering plan,
even across LANs and WANs.
Internetworking 247

Internetworking 248

Internetworking 249

Internetworking 250

Internetworking 251
Routing Algorithms
 In most subnets packets have to take

multiple hops
 the algorithms that choose the routes and the
data structures that they use are a major
area of network layer design
 the routing algorithm that is a part of the
network layer software responsible for
deciding which output line an incoming
packet should be transmitted on
 In session routing route remains in force for
an entire user session

Internetworking 252
Routing Algorithms
 Characteristics  Decision Time and Place

required  Time
 Correctness • Packet or virtual circuit
 Simplicity basis
 Robustness
 Place
• Distributed
 Stability
• Made by each node
 Fairness • Centralized
 Optimality • Source
 Efficiency
 Routing Strategies
 Performance Criteria  Non-adaptive
 Used for selection of route • Flooding
 Minimum hop • Random
 Least cost  Adaptive

Internetworking 253
Advanced Java Programming Flooding

 No network info required
 Packets are sent by source node to every
neighbor
 Incoming packets are retransmitted on every
link except incoming link
 Eventually a number of copies will arrive at
destination

Internetworking 254
Advanced Java ProgrammingFlooding Example

Internetworking 255
Properties of Flooding
 All possible routes are tried

 Very robust
 At least one packet will have taken minimum
hop count route
 Can be used to set up virtual circuit
 All nodes are visited
 Useful to distribute information (e.g. routing)

Internetworking 256
Disadvantage of Flooding
 Huge number of packets will be generated

 Solution
 Each packet is uniquely numbered so duplicates can be
discarded
 Nodes can remember packets already forwarded to keep
network load in bounds
 Can include a hop count in packets

Internetworking 257
Advanced Java Programming Random Routing

 Node selects one outgoing path for
retransmission of incoming packet
 Selection can be random or round robin
 Can select outgoing path based on probability
calculation
 No network info needed
 Route is typically not least cost nor minimum
hop

Internetworking 258
Advanced Java ProgrammingAdaptive Routing

 Used by almost all packet switching networks
 Routing decisions change as conditions on the
network change
 Failure
 Congestion
 Requires info about network
 Decisions more complex
 Tradeoff between quality of network info and
overhead
 Reacting too quickly can cause oscillation
 Too slowly to be relevant

Internetworking 259
Adaptive Routing - Advantages

 Improved performance
 Aid congestion control (See chapter 12)
 Complex system
 May not realize theoretical benefits

Internetworking 260
Advanced Java Programming Classification

 Based on information sources
 Local (isolated)
• Route to outgoing link with shortest queue
• Can include bias for each destination
• Rarely used - do not make use of easily available info
 Adjacent nodes
 All nodes

Internetworking 261
Isolated Adaptive Routing


Internetworking 262
ARPANET Routing Strategies(1)

 First Generation
 1969
 Distributed adaptive
 Estimated delay as performance criterion
 Bellman-Ford algorithm (appendix 10a)
 Node exchanges delay vector with neighbors
 Update routing table based on incoming info
 Doesn't consider line speed, just queue length
 Queue length not a good measurement of delay
 Responds slowly to congestion

Internetworking 263

 Second Generation
 1979
 Uses delay as performance criterion
 Delay measured directly
 Uses Dijkstra’s algorithm (appendix 10a)
 Good under light and medium loads
 Under heavy loads, little correlation between reported delays
and those experienced

Internetworking 264

 Third Generation
 1987
 Link cost calculations changed
 Measure average delay over last 10 seconds
 Normalize based on current value and previous results

Internetworking 265

Internetworking 266

Internetworking 267

Internetworking 268
1
Mean delay, T= µC −λ

Internetworking 269

Internetworking 270
Distance Vector Routing


Internetworking 271
Count to Infinity Problem


Internetworking 272

Internetworking 273

Internetworking 274

Internetworking 275
Advanced Java Programming Link State Routing

Internetworking 276

Internetworking 277
Advanced Java ProgrammingHierarchical Routing

Internetworking 278

Internetworking 279

Internetworking 280

Internetworking 281
Congestion Control
 when too many packets are present in (a part of the) subnet,

performance degrades, and the situation is called congestion
 as traffic increases too far, the routers are no longer able to
cope, and they begin losing packets
 at very high traffic the performance collapses completely, and
almost no packets are delivered
 congestion can be brought about by insufficient memory to
hold the packets, slow processors, low-bandwidth lines
 Congestion tends to feed upon itself and become worse
 congestion control has to do with making sure the subnet is
able to carry the offered traffic, flow control, in contrast
relates to point-to-point traffic between a given sender and a
receiver

Internetworking 282
Advanced Java Programming Congestion

Internetworking 283
General Principles
 control theory approach divides the solutions into two

groups: open loop and closed loop
 open loop solutions tend to solve the problem by good
design, in essence, to make sure that it does not occur
in the first place
 closed loop solutions are based on the concept of
feedback loop
 closed loop handles congestion control by:
 monitoring the system to detect when and where
congestion occurs
 passing this information to place where action can
be taken
 adjusting system operations to correct the problems
 open loop manages congestion by a technique called
traffic shaping - forcing the packets to be transmitted
©at a more
Copyright 2007, predictable
U. K. Roy, IT, JU rate 10/18/08 18:26
Internetworking 284

Internetworking 285
Leaky Bucket Algorithm

•Traffic shaping is used to regulate the average rate of data

transmission which is implemented using the leaky bucket algorithm
and the token bucket algorithm
Internetworking 286

Internetworking 287
Token Bucket Algorithm


Internetworking 288

Internetworking 289
Congestion Control in Virtual Circuit


Internetworking 290
Advanced Java Programming Waiting Fair Queue

Internetworking 291
Advanced Java Programming Chock Packet

Internetworking 292
Advanced Java Programming IP Addressing
 IP Address Requirements
 What is an IP Address?
 Network IDs and Host IDs
 What is a Physical Segment?
 IP Addressing Rules
 Classfull IP Addressing
 Address Classes
 Class A Addresses
 Class B Addresses
 Class C Addresses
 Class D & E Addresses
 Address Class Summary

Internetworking 293

IP Addressing Requirements
 Each Device that uses TCP/IP needs at least

one!
 Computer/Host (each Network Interface Card)
 Routers (each port or connection)
 Printers

Other Devices
 Each Device needs a Unique IP Address
 An Example:

206.77.105.9
 Configured in TCP/IP Software

Internetworking 294
Advanced Java Programming What is an IP Address?
 32-bit Binary Number (Address)

 11000000101010000111000100010011
 Divided into 4, 8-bit Octets
 11000000.10101000.01110001.00010011
 Converted to Decimal Numbers
 See: Binary Math
 192.168.113.19

Decimal range of an Octet: 0-255
 It contains the device’s:
 Network ID and Host ID

Internetworking 295
Advanced Java Programming Network ID and Host ID
 Network ID  Host ID
 Shared or Common to all  Identifies a specific device
computers on the same (Host) within a physical
physical segment
segment
 Unique on the Entire
Network
 Unique on the physical
segment
 “Area Code”
 “Phone Number”
192.176.11.201

Internetworking 296

What is a Physical Segment?
 A Broadcast Domain
 The portion of the network that you can retrieve
information from by using a broadcast packet!
 Ignore Repeaters, Bridges, or Switches
 Forward Broadcasts
 Everything (all devices) --
 Out a port of a router
 Between two routers
 Routers Don’t Forward Broadcasts
 IP Addressing Rules
 All Devices on the Same Physical Segment Share a
Common Network ID
 Each Physical Segment Has a Unique Network IDs

Internetworking 297

IP Addressing Rules
 Each Device (Host) Needs at Least One

Unique IP Address All Devices on the
Same Physical Segment Share a
Common Network ID (Subnet Mask)
Each Physical Segment Has a Unique
Network ID (Subnet Mask)

Internetworking 298

Classfull IP Addressing
 Traditional Manner of Addressing

 Class A
 Class B
 Class C
 Address Classes Specify Which Octets of the IP
Address are the Network-ID and Which are the
Host-ID
 Address Classes Specify Network Sizes
(Number of Hosts)

Internetworking 299

Address Classes
 Class A
 Network . Host . Host . Host
 Class B
 Network . Network . Host . Host
 Class C
 Network . Network . Network . Host
 Class D & E

Internetworking 300

Class A Networks:
The Definition
 Per Specification:
 1st Octet is the Network ID
 2nd, 3rd, 4th Octets are the Host ID
 In Binary –
 Any address that starts with a “0” in the first bit!
 First Class A Network Address:

00000001.00000000.0000000.00000000 (Binary)

1.0.0.0 (Decimal)
 Last Class A Network Address:
 01111111.00000000.00000000.00000000 (Binary)
 127.0.0.0 (Decimal) (Loopback Address)

Internetworking 301

Class A Networks:
Network IDs
 1st Octet is the Network ID
 0.0.0.0 (Invalid)
 1.0.0.0
 2.0.0.0
 3.0.0.0
 ~~~~
 127.0.0.0 (Loop back)
 2nd, 3rd, 4th Octets are the Host IDs
 An Assigned Class A Network Address:
 33.0.0.0 (Specifies the Network)
 2nd, 3rd, 4th Octets are the Host IDs
 Specified by Network Administrators

Internetworking 302
IP Addressing
Class A Networks:
The Number of Networks
 1st Octet is the Network ID

 1-126 = 126 Possible Class A Network IDs
 2nd, 3rd, 4th Octets are the Host IDs
 Each of the three Octets has a possible 256 Host IDs
 Number of Host IDs from three Octets:
 256 * 256 * 256 = 16,777,216 (minus 2) = 16,777,214
 Always Subtract 2 from the number of Host IDs
 Host IDs cannot be all 1’s (reserved for broadcast
address)
 Host IDs cannot be all 0’s (reserved for “this network
only” address)

Internetworking 303

Class A Networks:
Host ID Addresses
 33.0.0.0 (An Assigned Class A Address)

 All devices would share the 33 network ID.
 The Administrator would number the IP devices:

33.0.0.1 – 33.0.0.255 (255 Addresses)
 33.0.1.0 – 33.0.1.255 (256 Addresses)
 ~~~~

33.0.255.0 -- 33.0.255.255 (256 Addresses)
 (A Total of 65,535 Addresses)
 33.1.0.0 -- 33.1.255.255 (65,536 Addresses)
 33.2.0.0 -- 33.2.255.255 (65,536 Addresses)
 ~~~~
 33.255.0.0 -- 33.255.255.254 (65,535 Addresses)
 ( Total Addresses: 16.7 Million)

Internetworking 304

Class B Networks:
The Definition
 1st and 2nd Octets are the Network ID
 3rd, 4th Octets are the Host IDs
 In Binary –
 Any address that starts with a “10” in the first two
bits of the first octet!
 First Class B Network Address:
 10000000.00000000.0000000.00000000 (Binary)
 128.0.0.0 (Decimal)
 Last Class B Network Address:

10111111.11111111.00000000.00000000 (Binary)

191.255.0.0 (Decimal)

Internetworking 305

Class B Networks:
Network IDs
 1st and 2nd Octets are the Network IDs

128.0.0.0
 128.1.0.0
 ~~~~
 128.255.0.0
 129.0.0.0
 129.1.0.0
 ~~~~
 191.255.0.0
 3rd, 4th Octets are the Host IDs
 An Assigned Class B Network Addresses
 153.11.0.0
 3rd, 4th Octets are the Host IDs

Specified by Network Administrators

Internetworking 306

Class B Networks:
 1st and 2nd Octets are the Network IDs

 1st Octet 128 -- 191 = 64 Possible Network IDs
 2nd Octet 0 – 255 = 256 Possible Network IDs
 Total Class B Network IDs 64 * 256 = 16,384
 3rd, 4th Octets are the Host IDs
 Each of the Two Octets has a possible 256 Host IDs
 Number of Host IDs from Two Octets:
 256 * 256 = 65,536 (minus 2) = 65,534
 Host ID cannot be all 1’s (reserved for broadcast address)
 Host ID cannot be all 0’s (reserved for “this network only”
address)

Internetworking 307

Class B Networks:
Host ID Addresses
 An Assigned Class B Address

 153.11.0.0

All devices would share the 153.11 Network ID.

The Administrator would number the IP devices:
 153.11.0.1 -- 153.11.0.255 (255 Addresses)
 153.11.1.0 -- 153.11.1.255 (256 Addresses)
 153.11.2.0 -- 153.11.2.255 (256 Addresses)
 ~~~~
 153.11.255.0 -- 153.11.255.254 (255 Addresses)

Total Addresses: 65,534

Internetworking 308

Class C Networks:
The Definition
 1st, 2nd, 3rd Octets are the Network ID
 4th Octet is the Host ID
 In Binary –
 Any address that starts with a “110” in the first three
bits of the first octet!
 First Class C Network Address:
 11000000.00000000.0000000.00000000 (Binary)
 192.0.0.0 (Decimal)
 Last Class C Network Address:

11011111.11111111.11111111.00000000 (Binary)

223.255.255.0 (Decimal)

Internetworking 309

Class C Networks:
Network IDs
 1st, 2nd, 3rd Octets are the Network IDs

 192.0.0.0 – 192.0.255. 0
 192.1.0.0 – 192.1.255.0
 ~~~~
 192.255.0.0 – 192.255.255.0
 193.0.0.0 – 193.255.255.0
 ~~~~
 223.0.0.0 – 223.255.255.0
 4th Octet is the Host IDs
 An Assigned Class C Network Address
 201.11.206.0
 4th Octet is the Host IDs
 Specified by Network Administrators

Internetworking 310

Class C Networks:
 1st, 2nd, 3rd Octets are the Network IDs

 1st Octet 192 -- 223 = 31 Possible IDs
 2nd Octet 0 – 255 = 256 Possible IDs
 3nd Octet 0 – 255 = 256 Possible IDs
 Total Class C Network IDs 32 * 256 *256 = 2,097,152
 4th Octet is the Host ID
 An Octet has a possible 256 IDs
 Number of Host IDs an Octet:
 256 (minus 2) = 254
 Host ID cannot be all 1’s (reserved for broadcast address)

Host ID cannot be all 0’s (reserved for “this network only”
address)

Internetworking 311

Class C Networks:
Host ID Addresses
 An Assigned Class C Address

 201.11.206.0

All devices would share the 201.11.206.0 Network
ID.

The Administrator would number the IP devices:
 201.11.206.1, 201.11.206.2, ~~~~ 201.11.206.254


Internetworking 312

Class D & E
 Class D
 Used by Multicast Applications
 Shared Addresses
 224.0.0.0 – 239.255.255.255
 Class E
 Experimental
 240.0.0.0 +

Internetworking 313

Address Classes:
Network IDs and Host IDs
 Class A (1st Octet 1-127)

 Network.Host.Host.Host
 Class B (1st Octet 128-191)
 Network.Network.Host.Host
 Class C (1st Octet 192-223)
 Network.Network.Network.Host

Internetworking 314

Address Class Summary
 1st Networks Hosts

IDs Octet IDs /Network
 Class A 1-127 126 16,777,214
 Class B 128-191 16,384 65,534
 Class C 192-223 2,097,152 254

Internetworking 315

IP Address: What is It?
 32-bit Binary Number (Address)

 11000000101010001110000100010011
 Divided into 4, 8-bit Octets
 11000000.10101000.11100001.00010011
 Converted to Decimal Numbers

See: Binary Math
 192.168.225.19
 Decimal range of an Octet: 0-255
 It contains the device’s:
 Network ID and Host ID

Internetworking 316

IP Addressing Rules
 Each Device (Host) Needs at Least One

Unique IP Address
 All Devices on the Same Physical
Segment Share a Common Network ID
(Subnet Mask)
 Each Physical Segment Has a Unique
Network ID (Subnet Mask)

Internetworking 317

Address Classes: Network IDs and Host
IDs
 Class A (1st Octet 1-127)

 Network.Host.Host.Host
 Class B (1st Octet 128-191)
 Network.Network.Host.Host
 Class C (1st Octet 192-223)
 Network.Network.Network.Host

Internet Working

Transféré par

Informations du document

Description originale:

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Internet Working

Transféré par

Droits d'auteur :

Formats disponibles

INTERNETWORKING

Dept. of Information Technology,

Advanced Java Programming Books

• William Stallings, “Data & Computer Communications”, Prentice Hall

• Alberto Leon Garcia & Indra Widjaja, “Communication Networks”, Tata

• Krouse, “Computer Networks”, Pearson publication

• Farouzan, “Computer Networks”

• Farouzan, “TCP/IP Protocol Suit”

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming Introduction

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advantages of Computer Networks

Advantages of Computer Networks

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Key Communications Tasks

 Transmission System Utilization

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Key Communications Tasks

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Key Communications Tasks

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Simplified Data Communications Model

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming Protocols

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming Protocols

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming OSI Model

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming OSI Model

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming OSI Model

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

TCP/IP Protocol Architecture

 Developed by the US Defense Advanced

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming OSI v TCP/IP

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

TCP/IP Protocol Architecture Model

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Frequency, Spectrum and Bandwidth

 Time domain concepts

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Continuous & Discrete Signals

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming Periodic Signals

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming Sine Wave

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Varying Sine Waves

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Advanced Java Programming Wavelength

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Frequency Domain Concepts

 Signal usually made up of many frequencies

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Addition of Frequency Components

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23

Spectrum & Bandwidth

© Copyright 2007, U. K. Roy, IT, JU 10/18/08 18:23