FUNDAMENTAL OF

COMPUTER NETWORKS
CONCEPTS OF COMMUNICATION AND NETWORKING

Basics of Data Communication

1
2. Accuracy: The system must deliver data
Communication can be defined as the exchange of accurately. Data that have been altered in transmission
information of any kind by any means from one location and left uncorrected are unusable.
3. Timeliness: The system must deliver data in a
to another location. Communication system is the system
timely manner. Data delivered late are useless. In the
that is involved to achieve the goal of information
case of video and audio data timely deliver means
exchange. For example, exchange of voice signals
delivering data as they are produced, in the same order
between two telephones over the same network.
that they are produced without significant delay. This
Communication between a workstation and a server over kind of delivery is called real time transmission.
a public telephone network, so data communication is A data communication system is made up of five
the transfer of digital or analog data using digital or components:
analog signals over some form of network.
In this chapter, we will discuss about the basic
concepts of data communication and their modes. How
to transmits the data from one location to another
location? We will also discuss the basic Network,
protocols and standards, and their designing issues.

CONCEPT OF COMMUNICATION SYSTEM

Data communication is the exchange of data C in
the form of os and is) between two devices via some
form of transmission medium (such as a network cables). 1. Message: The message is the information to be
Data communication is considered local if the communicated. It can consist of text, number, pictures,
communicating devices are in the same building or a sound or video or any combination of these.
similarly restricted geographical area, and is considered 2. Sender: The sender is the device that sends the
remote if the devices are farther apart. data message.
The effectiveness of a data communication system 3. Receiver: The receiver is the device that receives
depends on three fundamental characteristics. the message.
1. Delivery: The system must deliver data to the 4. Medium: The transmission medium is the
correct destination. Data must be received by the physical path by which a message travels from sender
intended device or user and only by that device or user. to receiver.
2 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

5. Protocal: A Protocal is a set of rules that govem

data communication. A SAMPLE OF ANALOG
ANALOG AND DIGITAL COMMUNICATION
Sound
Analog Data Transmission: Sound coming out from
level
an instrument is an analog data communication. Analog data
is continuous over an interval. A representation of analog
time
signal is shown in the Figure.
Another example of analog data is video.
Digital Data Transmission: Digitial data is discrete. Examples are text and integers. They have to be
represented by a sequence of bits for communication. A sample is shown below:
Binary Code

Character
strings
from a
text

Particulars Analog Digital Transmission

1. Form Continuous variable of Discrete quantities
physical quantities, such with binary digits.
as electric current.
2. Cost of transmission Low High
3. Efficiency Low High
4. Maintenance cost High Low
5. Effect of noise High Low
6. Attenuation High Low
7. Examples Radio transmission, Data transmission
TV channels of text and integers

Comparison of Analog and Digital Data Communication: One drawback of analog communication is that
it is very sensitive to disturbances. Digital data communication has no such problem.
Analog and digital are the two types of signals used in data transmission.
Analog Signal
Values of analog signals keep varying. The change in the values are used to represent data. The analog signal
is continuous and can be represented as sine wave. Examples of analog signal are:
Human voice, Video, Audio and music etc.
Y

Give
amplitude wave

X
O

Analog Signal
 BASICS OF DATA COMMUNICATION / 3

The analog signals vary in amplitude (Volume) and

frequency (pitch). These signals suffer from problem one bit
of attenuation on their way, and amplifiers have to be
High .......
used to overcome it. The disadvantage is amplifiers also
amplify noise along with the original signal. Therefore, Low
0 1 0 1 1 0 1 1 0
if a signal is distorted, it cannot be reconstructed and
becomes a permanent loss. Thus, analog transmission
Digital Data Transmissions
is not used when high level of accuracy of data
transmission is required. It is used in the telephone
system, where a small distortion of human As the digital signal travels along a medium (with
communication does not matter. limited bandwidth) it gets distorted, and therefore, after
Analog techniques have the unique capability to a certain distance, the signal becomes unrecognizable.
capture real world as it exists. But their perfect copying Before this happens, a hardware device called repeater
or reproduction is not possible. This is because analog is used to regenerate the digital signal. Exactly same
signals cannot be copied perfectly. signal is regenerated and there is no loss of information.
Therefore, the original analog signals are converted The number of repeaters depend on the distance between
into digital signals to preserve audio/video in orginal the source and the destination. The line between the
for indefinite time or preservation. two is called digital line. When analog information,
In the digital form, the original analog signals can music, voice, and video are converted into binary
be stored and transmitted without loss of integrity due digital form, they can be electronically manipulated,
to accuracy inherent in digital signals. preserved and regenerated perfectly at very high
Digital Data Transmission speeds. Any number of copies can be made in exactly
Digital data is discontinuous and discrete. All same quality. This is the greatest advantage of digital
computers are digital machines, which can work only processing.
at two levels of signals 0 and 1, or ON and OFF. All COMMUNICATION MODES
data for use in computer is digital encoded as a sequence In sending and receiving messages of data from
of 0’s and 1’s in series. one place to another many elements work together.
The output of computers is in the form of digital These elements put together called a system. The
signals. The bandwidth of a digital signal is material communication system passes data or information in
compared to any medium, which has a limited the most effective manner. A communication system is
bandwidth. shown in the Figure below:

TELEPHONE
LINE
M M
DE DE
MO MO

Communication system is made up of hardware, used is called modulation. The reverse process is
software and data transfer links. It transfers data in a conversion of analog signal to digital form at the
cost-effective manner across the links. destination is called demodulation. The process of
A communication system can be either analog or modulation and demodulation is done by a special
digital (or a combination of both). That is, the device called modem (short form for modulation/
information can be transmitted in either analog or digital demodulation).
form within the communication networks. For example, The analog form of message is sent via telephone
computer generated data is digital, whereas the line to the destination, where it is again converted into
telephone lines are convenient to carry analog signals. digital pulses by the modem at receiver computer. Thus
Digital data are sent over analog telephone lines by the message or data is transmitted and received by the
converting digital signals to analog form. The technique two computers.
4 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

As the analog signal passes through the telephone Thus the time interval between events A and B in Figure
lines, its quality descreases with distance. Amplifiers below is not the same as B and C.
are used to boost the strength of the signal. But the
amplifiers add noise to the signal and sometimes cause
disruption of the message. But good quality of modems
at both ends are able to minimize the effect of noise. Synchronous
Advantages of Digital Transmission over Analog Transmission

Transmission
Voice data, music and images (e.g. television,
fax and video) can be interspersed for more
efficient use of the circuits and equipment.
Much higher data rates are possible using Asynchronous Transmission: Asynchronous
existing telephone lines. transmission is often referred to as start-top
Digital transmission is cheaper than analog transmission because of the nature, that is, the sender
transmission, since it is not necessary to can send a character at any time convenient and the
accurately reproduce an analog waveform receiver will accept it. Asynchronous communication
passing through several amplifiers on a long lines remain in an idle state until the hardware on the
distance line. Being able to correctly line is ready to transmit. Since the line is idle, series
distinguish 0 from 1 is enough. of bits are sent to the receiving node to tell it about
Maintenance of a digital system is easier than data coming. When data is finished, the node has to
maintenance of analog system. A transmitted be told that the transmission is complete. Hence, the
bit is either received correctly or not. It is STOP bits are sent. This pattern continues for the
simpler to track down the errors. duration of the time the link is operative. Thus, in
A digital signal can pass through any number asynchronous transmission data is transmitted
of regenerators (amplifiers in analog systems) character by character at irregular times.
with no loss in signal and, thus, long distances Synchronous Transmission: Synchronous
with no information loss. But analog signals devices do not use Start and Stop bits. Coordination
always face some information loss when between the sender and the receiver is handled
amplified, and this loss is accumulative. Digital differently. There are two channels—one for data and
transmission has error rate. other for link synchronization. The channel for
Data Communication Modes: In a digital synchronization uses the integral clock in the
communication system, there are two methods for data hardware for link synchronization between the two
transfer: parallel and serial. Parallel connection have nodes when one of the nodes is ready to transmit data,
multiple wires running parallel to each other, and can a unique combination of bits called a synchronous
transmit data on all the wires simultaneously. Serial, on
character is sent to the receiver. Since the first
the other hand uses a single wire to transfer the data
character will probably get trashed, a second one
bits one at a time.
usually follows to ensure that synchronization is
Parallel Data: The parallel port on modern
computer systems is an example of a parallel complete.
communications connection. IDE hard-disk connection
are another good example of parallel connections in a ABCDEFGHIJKLMN
computer system.
Synchronous
Serial Data: The serial port on modern computers Transmission
is a good example of serial communications. USB and
fire wire are good examples of other serial
communication standards.
ASYNCHRONOUS AND SYNCHRONOUS
TRANSMISSION Synchronous transmission means blocking a group
Asynchronous refers to a series of events that take of characters in way records are blocked on magnetic
place which are not synchornized one after the other. tape.
 BASICS OF DATA COMMUNICATION / 5

Each block is framed by an header and trailer Distinction between Asynchronous

information. The header consists of synchronizing and Synchronous Transmission
information which is used by the receiving device to set Synchronous communications are more expensive
its clock in synchronism with the sending end clock. than asynchronous because of costly hardware (integral
The header also has information to identify sender and clocking mechanism). Yet, synchronous communi-
receiver. Following the header is a block of characters
cation does eliminate about 20% overheads inherent
that contains the actual message to be transmitted. The
in asynchronous communications. It also allows greater
number of characters may vary. The block is terminated
by a trailer. The trailer contains an end of message throughput of data and better error detection.
character followed by a check character to aid detection Synchronous transmission is well suited to remote
of any transmission error. Thus with synchronous communication between a computer and such devices
transmission entire blocks of characters are framed and as buffered card readers and printers. It is also used for
transmitted together. computer to computer communication.

MODES OF DATA TRANSMISSION

T he maj or advantage of synchronous ends. Therefore, synchronous equipment usually costs

transmission is its efficiency. It eliminates the need more.
for individual start-stop bits on each character, and Asynchronous transmission is well suited to
so much higher data rates are possible than with many keyboard type terminals. It does not require
asynchronous transmission. The period between any local storage at the terminal or the computer as
blocks is kept small and the block itself is sent at transmission takes place character by character. It is
cheaper to implement. The main disadvantage is that
nearly the maximum line speed. This ensures efficient
the transmission line is idle during the time interval
utilization of the transmission line. The main between transmitting characters. Even though less
disadvantage is the need for local buffer storage at efficient than synchronous transmission, it is also used
the two ends of the line to assemble blocks and also with devices such as card readers and printers to
the need for accurately synchronized clocks at both reduce cost.
6 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

SIMPLEX, HALF DUPLEX AND FULL communication in both directions is necessary.

DUPLEX COMMUNICATION Even for a “one-way” flow of information from
Data Transmission Modes a terminal to a computer, the system is designed
There are three modes of data transmission related to let the computer to signal the terminal that
to the three types of used circuits. They are: data has been received. Without this ability, the
Simplex Mode; remote user who enters data may never know
Half-duplex Mode; and that it was not received by the other terminal.
Full-duplex Mode. Simplex circuits are enough, thus seldom used
Simplex Mode: Simplex communication is a because a return path is needed to send an
simple method of communicating in which there acknowledgement or control error signals.
is a one-way communication. Television Data Transmission
transmission is an example of simplex in Half-Duplex Mode
communications. The main transmitter sends out In half-duplex mode, both units communicate over
a signal (broadcast), but does not expect a reply the same medium, but only one can use the line at a
as the receiving units cannot send a reply back time. While one is in send mode, the other unit is in
to the transmitter. Another example of simplex receive mode. It is like two persons talking to each other
communication is a keyboard of a computer, one talk and the other listens, but neither one talks at
because the keyboard can only send data to the the same time. A half duplex line can, thus, alternately
computer. send and receive data. It requires two lines. This is the
most common type of transmission for voice
Data-flow communication because only one person is supposed
Transmitter Receiver to speak at a time. It is also used to connect a terminal
Simplex configuration on with a computer. The terminal might transmit data and
For many types of application simplex is good then the computer responds with an acknowledgement.
enough to flow the information unidirectionaly. The transmission of data to and from a hard disk is also
However, in data processing applications, an half-duplex mode transmission.

Transmitter and Receiver Receiver and Transmitter

Half-duplex configuration

Full-Duplex System of Data unacceptable in some cases. Also many applications

Transmission require simultaneous transmission in both directions.
In a half-duplex system the line has to be “turned In such cases, a full-duplex system is used that permits
around” every time the direction of transmission is information to flow simultaneously in both directions
reversed. This needs a special switching circuit and a on the transmission path. Full-duplex line improves
small amout of time (approximately 150 milliseconds). efficiency as the line turn-around time required in a half-
With high speed computers, this turn-around time is duplex arrangement is saved. It requires four wires.

Transmitter and Receiver Receiver and Transmitter

Full-duplex configuration
 BASICS OF DATA COMMUNICATION / 7

Difference between Simplex, Half-duplex and Full-duplex modes of Data Transmission

Type Simplex Half-Duplex Full-Duplex
1. Direction of Communication in one direction Both ways, but Communication in both
transmission only communication directions simultaneously
in one direction
at a time
2. Confirmation No confirmation Delayed Immediate
of the comm- confirmation confirmation
unication
3. Number of Two Two Four
wires required
4. Cost of trans- Low Average High Data
mission transmission
5. Examples of T.V. and Radio transmission and Telephonic talk
users broadcasting
6. Efficiency Low Medium High

NETWORKING PROTOCOLS vendors, government agencies and other service

AND STANDARDS providers to ensure the kind of inter-connectivity
In computer network communication occurs necessary in today’s market place and in international
between entities in different systems. An entity is communications.
anything capable of sending or receiving information. Data communication standards fall into two
Examples include applications programmes, file transfer category:
packages, browser data base management systems and (i) de facto’ (meaning “by fact”) and
electronic mail software. A system is a physical object (ii) de jure (meaning “by law”)
that contains one or more entities, examples include De facto standards can be further subdivided into
two classes: proprietary and non-proprietary. Proprietary
computers and terminals.
standards are those originally invented by a commercial
But two entities cannot just send bit stream to each
organization as a basic for the operation of its products.
other and expect to be understood. For communication Non-proprietary standards are those originally
to occurs, the entities must agree on a protocol. A developed by groups or committees that have passed
protocol is a set of rules that govern data communication. them into the public domain. It is also called open
A protocal defines what is communicated, how it is standards, because they open communications between
communicated and when it is communicated. The key different systems.
elements of a protocol are syntax, semantics and timing. De jure standards are those that have been legislated
Syntax: It refers to the structure or format of the by an officially recognized body. Standards that have
data, meaning the order in which they are presented. not been approved by an organized body but have been
Semantics: Semantics refers to the meaning of each adopted as standard through wide-spread use are de facto
section of bits. standards.
Timing: Timing refers to two characteristics: when Standards are developed by cooperation among
standards creation committees forums, and government
data should be sent and how fast they can be sent.
regulatory agencies.
Standards: A standard provides a model for ISO (International Standards Organization)
development that makes it possible for a product to work ITU-T (International Telecommunication Union
regardless of the individual manufacturer. Standards are Telecommunication Standards Sector)
essential in creating and maintaining an open and ANSI (American National Standard Institute)
competitive market for equipment manufacturers and IEEE (Institute of Electrical and Electronics
in guaranteeing national and international inter- Engineers)
operability of data and telecommunications technology EIA (Electronic Industries Association).
and processes. They provide guidelines to manufacturer, Telecordia.
8 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

LAYERING rules, and data formats specified by protocols. The blue

Many communications protocols can be viewed as lines therefore, mark the boundaries of the (horizontal)
layered protocols where each layer gives a service to protocol layers.
the protocol of the layers above and requires service The vertical protocols are not layered because they
from layers below. don’t obey the protocol layering principle which states
Typically, a (best-effort) hardware delivery that a layered protocol is designed so that layer at the
mechanism layer is used to build a connectionless packet destination receives exactly the same object sent by layer
delivery system on top of which a reliable transport layer at the source. The horizontal protocols are layered
is built, on top of which is the application software. protocols and all belong to the protocol suite. Layered
Layers below and above can be defined, and protocols protocols allow the protocol designer to concentrate on
are very often stacked to give tunnelling, for example, one layer at a time, without worrying about how other
the internet protocol can be tunnelled across an ATM layers perform.
network protocol to provide connectivity by layering The vertical protocols need not be the same
the internet protocol on top of the ATM protocol protocols on both systems, but they have to satisfy some
transport layer. minimal assumptions to ensure the protocol layering
The delivery system is defined by the IP protocol principle holds for the layered protocols. This can be
and the transport system by the TCP protocol. achieved using a technique called Encapsulation.
Protocol Layering Usually, a message or a stream of data is divided
into small pieces, called messages or streams, packets,
IP datagrams or network frames depending on the layer
in which the pieces are to be transmitted. The pieces
contain a header area and a data area. The data in the
header area identifies the source and the destination on
the network of the packet, the protocol, and other data
meaningful to the protocol like CRC’s of the data to be
send, data length, and a timestamp.
The rule enforced by the vertical protocols is that
the pieces for transmission are to be encapsulated in
the data area of all lower protocols on the sending side
Message flows using a protocol suite.

Protocol layering now forms the basis of protocol and the reverse is to happen on the receiving side. The
design. It allows the decomposition of single, complex result is that at the lowest level the piece looks like this:
protocols into simpler, cooperating protocols, but it is ‘Header1, Header2, Header3, data’ and in the layer
also a functional decomposition, because each protocol directly above it: ‘Header2, Header3, data’ and in the
belongs to a functional class, called a protocol layer. top layer: ‘Header3,data’, both on the sending and
The protocol layers each solve a distinct class of receiving side. This rule therefore, ensures that the
communications problems. The Internet protocol suite protocol layering principle holds and effectively
consists of the following layers: application, transport- virtualizes all but the lowest transmission lines, so for
internet- and network interface-functions. Together, the this reason some message flows are coloured red in
layers make up a layering scheme or model. figure.
In computations, we have algorithms and data, and To ensure both sides use the same protocol, the
in communications, we have protocols and messages, pieces also carry data identifying the protocol in their
so the analog of a data flow diagram would be some header.
kind of message flow diagram. To visualize protocol The design of the protocol layering and the network
layering and protocol suites, a diagram of the message (or Internet) architecture are interrelated, so one cannot
flows in and between two systems, A and B, is shown in be designed without the other. Some of the more
figure above. important features in this respect of the Internet
The systems both make use of the same protocol architecture and the network services it provides are
suite. The vertical flows (and protocols) are in system described next.
and the horizontal message flows (and protocols) are The Internet offers universal interconnection,
between systems. The message flows are governed by which means that any pair of computers
 BASICS OF DATA COMMUNICATION / 9

connected to the Internet is allowed to address of a router known to be closer to the destination.
communicate. Each computer is identified by A special entry can specify that a default router is chosen
an address on the Internet. All the interconnected when there are no known paths.
physical networks appear to the user as a single All networks are treated equal. A LAN, a WAN
large network. This interconnection scheme is or a point-to-point link between two computers
called an internetwork or internet. are all considered as one network.
Conceptually, an Internet addresses consists of A connectionless packet delivery (or packet-
a netid and a hostid. The netid identifies a switched) system (or service) is offered by the
network and the hostid identifies a host. The Internet, because it adapts well to different
term host is misleading in that an individual hardware, including best-effort delivery
computer can have multiple network interfaces mechanisms like the Ethernet. Connectionless
each having its own Internet address. An Internet delivery means that the messages or streams are
Address identifies a connection to the network, divided in pieces that are multiplexed separately
not an individual computer. The netid is used on the high speed inter machine connections
by routers to decide where to send a packet. allowing the connections to be used
Network technology independence is achieved concurrently. Each piece carries information
using the low-level address resolution protocol identifying the destination. The delivery of
(ARP) which is used to map Internet addresses packets is said to be unreliable, because packets
to physical addresses. The mapping is called may be lost, duplicated, delayed or delivered
address resolution. This way physical addresses out of order without notice to the sender or
are only used by the protocols of the network receiver. Unreliability arises only when
interface layer. The TCP/IP protocols can make resources are exhausted or underlying networks
use of almost any underlying communication fail. The unreliable connectionless delivery
technology. system is defined by the Internet Protocol (IP).
The protocol also specifies the routing function,
which chooses a path over which data will be
send. It is also possible to use TCP/IP protocols
on connection-oriented systems. Connection-
oriented systems build up virtual circuits (paths
for exclusive use) between senders and
receivers. Once build up the IP datagrams are
send as if they were data through the virtual
circuits and forwarded (as data) to the IP
protocol modules. This technique, called
Physical networks are interconnected by routers. tunneling, can be used on X.25 networks and
Routers forward packets between interconnected ATM networks.
networks making it possible for hosts to reach hosts on A reliable stream transport service using the
other physical networks. The message flows between unreliable connectionless packet delivery
two communicating system A and B in the presence of service is defined by the transmission control
a router R are illustrated in figure above. Datagrams protocol (TCP). The services are layered as well
are passed from router to router until a router is reached and the application programs residing in the
that can deliver the datagram on a physically attached layer above it, called the application services,
network (called direct delivery). To decide whether a can make use of TCP. Programs wishing to
datagram is to be delivered directly or is to be send to a interact with the packet delivery system itself
router closer to the destination, a table called the IP can do so using the user datagram protocol
routing table is consulted. The table consists of pairs of (UDP).
network ids and the paths to be taken to reach known Software Layering
networks. The path can be an indication that the Having established the protocol layering and the
datagram should be delivered directly or it can be the protocols, the protocol designer can now resume with
10 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

the software design. The software has a layered application program passes data to the transport
organization and its relationship with protocol layering layer for delivery.
is visualized in figure below: Transport Layer: The transport layer provides
communication from one application to another.
The transport layer may regulate flow of
information and provide reliable transport,
ensuring that data arrives without error and in
sequence. To do so, the receiving side sends
back acknowledgments and the sending side
retransmits lost pieces called packets. The
stream of data is divided into packets by the
module and each packet is passed along with a
Message flows using a protocol suite. destination address to the next layer for
transmission. The layer must accept data from
The software modules implementing the protocols
many applications concurrently and therefore
are represented by cubes. The information flow between
also includes codes in the packet header to
the modules is represented by arrows. The (top two
identify the sending and receiving application
horizontal) red arrows are virtual. The blue lines mark
the layer boundaries. program.
To send a message on system A, the top module Internet Layer: The Internet layer handles the
interacts with the module directly below it and hands communication between machines. Packets to
over the message to be encapsulated. This module reacts be send are accepted from the transport layer
by encapsulating the message in its own data area and along with an identification of the receiving
filling in its header data in accordance with the protocol machine. The packets are encapsulated in IP
it implements and interacts with the module below it by datagrams and the datagram headers are filled.
handing over this newly formed message whenever A routing algorithm is used to determine if the
appropriate. The bottom module directly interacts with datagram should be delivered directly or send
the bottom module of system B, so the message is send to a router. The datagram is passed to the
across. On the receiving system B the reverse happens, appropriate network interface for transmission.
so ultimately (and assuming there were no transmission Incoming datagrams are checked for validity
errors or protocol violations etc.) the message gets and the routing algorithm is used to decide
delivered in its original form to the top module of system whether the datagram should be processed
above. locally or forwarded. If the datagram is
On protocol errors, a receiving module discards the addressed to the local machine, the datagram
piece it has received and reports back the error condition header is deleted and the appropriate transport
to the original source of the piece on the same layer by protocol for the packet is chosen. ICMP error
handing the error message down or in case of the bottom and control messages are handled as well in this
module sending it across. layer.
The division of the message or stream of data into Network Interface Layer: The network
pieces and the subsequent reassembly are handled in interface layer is responsible for accepting IP
the layer that introduced the division/reassembly. The datagrams and transmitting them over a specific
reassembly is done at the destination (i.e. not on any network. A network interface may consist of a
intermediate routers). device driver or a complex subsystem that uses
TCP/IP software is organized in four layers. its own data link protocol.
Application Layer: At the highest layer, the Program translation has been divided into four sub
services available across a TCP/IP internet are problems: compiler, assembler, link editor, and loader.
accessed by application programmes. The As a result, the translation software is layered as well,
application chooses the style of transport to be allowing the software layers to be designed
used which can be a sequence of individual independently. Noting that the ways to conquer the
messages or a continuous stream of bytes. The complexity of program translation could readily be
 BASICS OF DATA COMMUNICATION / 11

applied to protocols because of the analogy between split up into two protocols, at the cost of one or two
programming languages and protocols. The designers extra layers, but there is a hidden caveat, because the
of the TCP/IP protocol suite were keen on imposing the model is also used to provide a conceptual view on the
same layering on the software framework. This can be suite for the intended users. There is a trade-off to be
seen in the TCP/IP layering by considering the made here between preciseness for the designer and
translation of a pascal programme (message) that is clarity for the intended user.
compiled (function of the application layer) into an OSI REFERENCE MODEL
assembler program that is assembled (function of the The International Standards Organization (ISO)
transport layer) to object code (pieces) that is linked proposal for the Standardization of the various protocols
(function of the Internet layer) together with library used in computer networks specifically those networks
object code (routing table) by the link editor, producing used to connect open systems is called the Open
relocatable machine code (datagram) that is passed to Systems Interconnection Reference Model (1984), or
the loader which fills in the memory locations (ethernet simply the OSI Model
addresses) to produce executable code (network frame)
to be loaded (function of the network interface layer)
into physical memory (transmission medium). To show
just how closely the analogy fits, the terms between
parentheses in the previous sentence denote the relevant
analogs and the terms written cursively denote data
representations. Program translation forms a linear
sequence, because each layer’s output is passed as input
to the next layer. Furthermore, the translation process
involves multiple data representations. We see the same
thing happening in protocol software where multiple
protocols define the data representations of the data
passed between the software modules.
The network interface layer uses physical addresses
and all the other layers only use IP addresses. The
boundary between network interface layer and Internet
layer is called the high-level protocol address boundary. 1. Physical Layer: This layer describes the
The modules below the application layer are generally Electrical, mechanical and functional responsibility for
considered part of the operating system. Passing data movements of Individual bits from one node to the next
between these modules is much less expensive than node. Responsibilities of Physical layer include the type
passing data between an application program and the of transmission media used for transmitting the data from
transport layer. The boundary between application layer one node to another. Transmission medium defines the
and transport layer is called the operating system type of Transmission medium used for connecting two
boundary. devices for exchanging the information. These
Strict Layering transmission medium can be twisted pair wires, Coaxial
Strictly adhering to a layered model, a practice cable, Fibre optical Cable, Radio waves. Etc Physical
known as strict layering is not always the best approach layer encodes the Data bits into signals–electrically in
to networking. Strict layering can have a serious impact the form of “0” and “1”. Converting the data into
on the performance of the implementation, so there is different bit Stream is one of the responsibilities of the
at least a trade-off between simplicity and performance. physical layer. Physical layer also keeps record of that
Another, perhaps more important point can be shown how and in which format different layers are connected
by considering the fact that some of the protocols in the to each other. Whether they are connected Point to Point
Internet Protocol suite cannot be expressed using the or Multipoint.
TCP/IP model, in other words some of the protocols 2. Data Link Layer: The Data link layer is
behave in ways not described by the model. To improve responsible for moving frames from One node to the
on the model, an offending protocol could, perhaps be next on the Same network. A data frame is the organized,
12 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

logical structure in which data is placed. So Data link Responsibilities:

layer is responsible for error free transmission of data. (a) Translation: The Presentation layer is
Large Data is dividable into small manageable parts responsible for translating the information into sender
called Frame. The data link layer imposes flow control dependent format and at the receiving end into receiver
mechanism when the data sending speed is faster than dependent format.
the receiving speed so To prevent the data loss flow (b) Encryption: This layer is also responsible for
control is must. Sequence control is another data encryption/decryption technique. Encryption means
responsibility of Data link layer. transformation of original information into another form.
3. Network Layer: The Network layer responsible Decryption converts the encrypted information into
for the source to Destination delivery of packet in original form.
multiple networks .When two devices are connected to (c) Compression: Data compression reduces the
each other on the same network then we use data link number of bits to be transmitted. This information can
layer but when the two systems are on different A be in the form of text, audio or video. And decompress
network then there is a need for network layer. Network again brings the compressed information into Original
layer is responsible for appropriate routing of data. Other form. For e.g. WinZip, win rar etc.
responsibility of network layer is congestion control, 7 Application layer: Application provides user
accounting and problem of the heterogeneous networks, interfaces and support for services such as E-Mail,
sequence and flow control, resetting and releasing the remote file access and transfer shared database
connection. The devices used in network layer are router, management etc.
repeater, gateways etc. Responsibilities:
4. Transport Layer: Transport layer includes the (i) File transfer, access, management: It permits
function for establishing appropriate connection the user to access download or upload the file from
between the two devices, data transfer and connection remote computer.
release after the transmission is complete. The Transport (ii) Mail services: It permits the user to use mail
layer is responsible for the delivery of a message from services.
(iii) Remote login: It allows the host to login from
one process to another. The transport layer gets the entire
remote computer.
message to the correct Process on that computer. The
ENCAPSULATION
Network layer gets each packet to the Correct computer
The data unit created at the application layer is
and transport layer sends the Entire message to the
called a message. TCP DDP creates a data unit that is
correct process .packets are deliver error free and in
called either a segment or a user datagram. The IP layer
sequence with No loss or delicacy.
in turn creates a data unit called datagram. The
5. Session Layer: Session layer establishes, movement of the data gram across the internet is the
maintains, and synchronizes the Interaction among responsibility of the TCP/ IP protocol. However to be
Communicating systems. Session layer is responsible able to move physically from one network to another,
for establishing the connection before, sending the data, the datagram must be encapsulated in a frame in the
maintaining the connection till the data is transferred data link layer of the underlying network and finally
and Terminating sessions when the data is completely transmitted as signals along the transmission media.
transferred responsibility of Session layer includes the The physical address is associated with the data
name recognition of the two computers. Security link layer. The logical address is associated with the
required over the network, synchronization between the data link layer. The logical address is associated with
task by adding the check points between the task and the network layer. A port address is associated with the
dialog control is also of the responsibility of Session application layer.
layer. END-TO-END ARGUMENT
6. Presentation Layer: The Presentation layer is The end-to-end principle is a classic design
responsible for translation, compression, and encryption. principle of computer networking which states that
It means for the security of the data presentation layer application specific functions ought to reside in the end
provides services of encoding and decoding character hosts of a network rather than in intermediary nodes,
Sets. provided they can be implemented “completely and
 BASICS OF DATA COMMUNICATION / 13

correctly” in the end hosts. A common misperception messages in or between computing systems and in
about the end-to-end principle is that it implies network telecommunications. A protocol may have a formal
neutrality or a “dumb” network. First, there are functions description. Protocols may include signalling,
orthogonal to application specifics or which simply authentication and error detection and correction
cannot be implemented “completely and correctly” in capabilities. A protocol definition defines the syntax,
the network ends; and, second, there may be good semantics, and synchronization of communication; the
reasons to offload functions from end hosts to specified behaviour is typically independent of how it
intermediary nodes, for performance reasons or when is to be implemented. A protocol can therefore be
these are better equipped to perform the functions or implemented as hardware or software or both.
part thereof. The communications protocols in use on the
Some view the end-to-end principle as a pragmatic Internet are designed to function in very complex and
engineering philosophy for network system design that diverse settings. To ease design, communications
merely prefers putting intelligence towards the end protocols are structured using a layering scheme as a
points. It does not forbid intelligence in the network basis. Instead of using a single universal protocol to
itself if it makes more practical sense. From the handle all transmission tasks, a set of cooperating
beginning, the end-to-end arguments revolved around protocols fitting the layering scheme is used.
requirements that could be implemented correctly at the The layering scheme in use on the Internet is called
end-points; if implementation inside the network is the the TCP/IP model. The actual protocols are collectively
only way to accomplish the requirement, then an end to called the Internet protocol suite. The group responsible
end argument isn’t appropriate in the first place. for this design is called the Internet Engineering Task
In the Internet Protocol Suite, the Internet Protocol Force (IETF).
is a simple (“dumb”), stateless protocol that moves Obviously the number of layers of a layering
datagrams across the network, and TCP is a smart scheme and the way the layers are defined can have a
transport protocol providing error detection, drastic impact on the protocols involved. This is where
retransmission, congestion control, and flow control the analogies come into play for the TCP/IP model,
end-to-end. The network itself (the routers) needs only because the designers of TCP/IP employed the same
to support the simple, lightweight IP; the endpoints run techniques used to conquer the complexity of
the heavier TCP on top of it when needed. programming language compilers (design by analogy)
A second canonical example is that of file transfer. in the implementation of its protocols and its layering
Every reliable file transfer protocol and file transfer scheme.
programme should contain a checksum, which is Basic Requirements of Protocols
validated only after everything has been successfully Messages are sent and received on communicating
stored on disk. Disk errors, router errors, and file transfer systems to establish communications. Protocols should
therefore specify rules governing the transmission. In
software errors make an end-to-end checksum necessary.
general, much of the following should be addressed:
Therefore, there is a limit to how secure TCP checksum
Data Formats for Data Exchange: Digital
should be, because it has to be re implemented for any
message bit strings are exchanged. The bit
robust end-to-end application to be secure.
strings are divided in fields and each field carries
A third example is the Ether Type Field of Ethernet.
information relevant to the protocol.
An Ethernet frame does not attempt to provide
Conceptually the bit string is divided into two
interpretation for the 16 bits of type. To add special
parts called the header area and the data area.
interpretation to some of these bits would reduce the The actual message is stored in the data area,
total number of Ether types, hurting the scalability of so the header area contains the fields with more
higher layer protocols, i.e. all higher layer protocols relevance to the protocol. Bit strings longer than
would pay a price for the benefit of just a few. Attempts the maximum transmission unit (MTU) are
to add elaborate interpretation (e.g. IEEE 802 SSAP/ divided in pieces of appropriate size.
DSAP) have generally been ignored by most network Address Formats for Data Exchange: Addre-
designs. sses are used to identify both the sender and the
PROTOCOL DESIGN ISSUES intended receiver(s). The addresses are stored
A communications protocol is a system of digital in the header area of the bit strings, allowing
message formats and rules for exchanging those the receivers to determine whether the bit strings
14 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

are intended for themselves and should be direction at a time as on half-duplex links. This
processed or should be ignored. A connection is known as Media Access Control.
between a sender and a receiver can be Arrangements have to be made to accommodate
identified using an address pair (sender address, the case when two parties want to gain control
receiver address). Usually some address values at the same time.
have special meanings. An all-1s address could Sequence Control: We have seen that long bit
be taken to mean an addressing of all stations strings are divided in pieces, and then sent on
on the network, so sending to this address would
the network individually. The pieces may get
result in a broadcast on the local network. The
lost or delayed or take different routes to their
rules describing the meanings of the address
destination on some types of networks. As a
value are collectively called an addressing
result pieces may arrive out of sequence.
scheme.
Retransmissions can result duplicate pieces. By
Address Mapping: Sometimes protocols need
to map addresses of one scheme on addresses marking the pieces with sequence information
of another scheme. For instance, to translate a at the sender, the receiver can determine what
logical IP address specified by the application was lost or duplicated, ask for necessary
to an Ethernet hardware address. This is referred retransmissions and reassemble the original
to as address mapping. message.
Routing: When systems are not directly Flow control is needed when the sender
connected, intermediary systems along the route transmits faster than the receiver or intermediate
to the intended receiver(s) need to forward network equipment can process the trans-
messages on behalf of the sender. On the missions. Flow control can be implemented by
Internet, the networks are connected using messaging from receiver to sender.
routers. This way of connecting networks is APPLICATIONS
called internetworking. Data communication networks have today become
Detection of transmission errors is necessary on an integral part of business, industry, and entertainment.
networks which cannot guarantee error-free Some of the network applications in different fields are
operation. In a common approach, CRCs of the as follows:
data area are added to the end of packets, Marketing and Sales: Computer networks are
making it possible for the receiver to detect used by marketing professionals to collect,
differences caused by errors. The receiver exchange, and analyze data relating to
rejects the packets on CRC differences and customers needs and product development
arranges somehow for retransmission. cycles. Sales applications include teleshopping,
Acknowledgements of correct reception of and on-line reservation services for hotels,
packets are required for connection oriented airlines, etc.
communication. Acknowledgements are sent Financial Services: Financial services are now
from receivers back to their respective senders. more and more dependent on computer
Loss of information-timeouts and retries. networks. Applications include credit history
Packets may be lost on the network or suffer searches; foreign exchange and investment
from long delays. To cope with this, under some services, and electronic funds transfer (EFT).
protocols, a sender may expect an Manufacturing: Computer networks are used
acknowledgement of correct reception from the in many aspects of manufacturing, including
receiver within a certain amount of time. On the manufacturing process itself. Two such
timeouts, the sender must assume the packet was applications are computer-assisted design
not received and retransmit it. In case of a (CAD) and computer-assisted manufacturing
permanently broken link, the retransmission has (CAM).
no effect so the number of retransmissions is Electronic Messaging: It is the most widely
limited. Exceeding the retry limit is considered used network application.
an error. Directory Services: Directory services allow
Direction of information flow needs to be lists of files to be stored in a central location to
addressed if transmissions can only occur in one speed worldwide search operations.
 BASICS OF DATA COMMUNICATION / 15

Information Services: Network information require less power but (typically more bandwidth than
services include bulletin boards and data banks. analog and copies of digital signals can be made exactly
A World Wide Web site transfers for a new like the original.
product is an information service. Q. 5. What is Analog signal?
Ans. A continuously varying signal or wave. As
Electronic Data Interchange (EDI): EDI
with all waves, analog waves are susceptible to
business information (including documents interference which can change the character of the wave.
such as purchase order, and invoices) without
using paper.
Teleconferencing: Teleconferencing allows
conferences to occur without the participants
being in the same place.
Cellular Telephone: Cellular networks make
it possible to maintain wireless phone
connections even while travelling.
Cable Television: Cable television networks
may soon provide video on request, as well as
the information, financial and communication Q. 6. What is data rate?
services currently provided by the telephone Ans. The maximum number of bits of information
companies and computer networks. which can be transmitted per seconds is data
transmission link. Typically expressed as mega bits per
second (mbps). ‘This shows the speed at which data
Q. 1. Explain the data transmission concept? can be transmitted between devices. This is sometimes
Ans. Data transmission is the moving of data in referred to as though put. it can also be expressed in
the form of electromagnetic signals across a transmission MBps.
medium. As an example, when you are collecting 1, 24 bits = 1 kb
numerical statistics from another computer, sending
8 kb = 1 kil byte
animated pictures from a design workstation, or causing
128 kb = 1 megabit
a bill to ring at a distant control center, at that movement
8 mb = 1 MB
you are facing data transmission.
1, 024 kB = 1 Megabyte
Q. 2. Explain the term Analog data transmission
128 MB = 1 gegabit
and Typically from 20Hz to 20KHz what is the
8 gb = 1 GB.
frequency range for audio signal?
1024 MB = 1 GB.
Ans. The method of transmission in which we use
Q. 7. Write disadvantages of analog signals?
or send analog data over transmission medium. As an
Ans. The most popular disadvantages of the analog
example, Analog data is human voice. When someone
system is random variation. As the signal is copied and
speaks, an analog wave is created in the air. This can be
recopied, or transmitted over long distance, these
captured by a microphone and converted to an analog
signal. random variations become dominant. Electrically, these
Q. 3. What is digital data transmission? losses can be diminished by shielding, good connections,
Ans. The method of transmission in which we use and several cable types such as coaxial or twisted pair.
or send digital data over transmission medium. The The effects of noise make signal loss and distortion
digital data is the data stored in the memory of the impossible to recover, since amplifying the signal to
computer in the form of 0s and 1s. It can be converted recover attenuated parts of signal amplifies the noise as
to digital signal where it is transferred from one position well. Even if the resolution of an analog signal is higher
to another inside or outside the computer or modulated than a comparable digital signal, in many cases, the
into an analog signal and there sent through a difference is overshadowed by the noise in the signal.
transmission medium to another computer. Q. 8. What are various types of Communication
Q. 4. What is digital signal? / Transmission?
Ans. A signal that takes on only two values, off or Ans. 1. Synchronous Transmission
on, typically represented by 0 and I. Digital signals 2. Asynchronous Transmission.
16 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Synchronous Transmission: When data is being Q. 9. What is serial and parallel transmission?
send at periodic intervals with constant arrival rate then Ans. Serial Transmission: In this transmission, the
data is said to be synchronized. Here, each data block data is transmitted serially over the line. As signal wire
is being surrounded by control bits called flags which pair can carry the signal so generally preferred for long
delimits the data block boundaries and helps the receiver distance communication. The problem of crosstalk,
in notifying the arrival of data. So here a separate clock interference is negligible but the speed is slow. Hence
or the clocking codes are being used because a drift in due to its low cost in implementation, optimum data
clock frequency can lead to loss of data. rate for longer distance it is considered efficient
transmission e.g. keyboard to computer system.

Synchronous data frame

Asynchronous Transmission: When a data Serial Transmission

character is sent at any time i.e. no fixed interval this Parallel Transmission: Here, the multiple data
can lead to loss of synchronizations as it is then difficult lines are laid out to carry multiple signals at a time. The
to know that when next character is arriving, hence in lines are laid in multiple of 8. With the distance cost
this approach, each character is being preceded by a bit gets increased as more number of cables are to be drawn,
called start bit while succeeded by a stop bit. The arrival hence seeing its cost factor this transmission is generally
of these bits will tell the receiver about the corresponding preferred for short distances communication. The
character. multiple paths also introduced the concept of cross
This type of communication is generally found in talking and interference leading to the generation of
machines or terminals such as teletypes or teleprinters errors. But it is fast as multiple bits are transmitted
and low speed computer terminals. It is very simple to simultaneously on different paths to the receiver. e.g.
implement.
data transmission between computer system and printer.

Asynchronous Transmission

In the above figure the varying line lengths show

that the arrival time of various characters is not fixed.
Generally in these systems, the start bit is placed as 0 Parallel Transmission
and stop bit is 1 because when no data transmission is
taking place the line is idle.
 17

Modulation and Encoding

2

Modulating Signal (audio)

A computer network is designed to send
information from one point in the network to another. In
designing a network, we have two choices. Convert
AM signal
information to either a digital signal or an analog signal. Carrier frequency

In this chapter, we discuss several techniques to convert

analog to digital and digital to analog.
In this chapter, we will also discuss the advantages Amplitude Modulation
and disadvantages of digital transmission over analog
Frequency Modulation (FM)
transmission. We will discuss some of the techniques
In FM transmission, the frequency of the carrier
and concepts which can be used to transmit analog and
signal is modulated to follow the changing voltage level
digital data.
(amplitude) of the modulating signal. The peak
amplitude and phase of the carrier signal remain
ANALOG MODULATION (AM, FM, PM): constant, but as the amplitude of the information signal
Analog-to-analog encoding is the representation of changes, the frequency of the carrier changes
analog information by an analog signal. Radio is an proportionately. Figure below shows the relationships
example of an analog-to-analog communication. Figure of the modulating signal, and the resultant FM signal.
below shows the relationship between the analog Amplitude Modulating signal (audio)
information, the analog-to-analog conversion hardware, Time
and the resultant analog signal.
Analog-to-analog encoding can be accomplished Amplitude Carrier frequency Amplitude FM signal
in three ways—amplitude modulation (AM), frequency Time Time
modulation (FM), and phase modulation (PM).
Wave

Analog/ FM Bandwidth
analog The bandwidth of an FM signal is equal to 10 times
encoding the bandwidth of the modulating signal and, like AM
Analog-to-analog encoding bandwidths, covers a range centred around the carrier
frequency.
Amplitude Modulation (AM)
Phase Modulation (PM)
In AM trans-mission, the carrier signal is modulated
Phase modulation (PM) is used sometimes as an
such that its amplitude varies with the changing
alternative to frequency modulation. In PM
amplitudes of the modulating signal. The frequency and transmission, the phase of the carrier signal is modulated
phase of the carrier remain the same; only the amplitude to follow the changing voltage level (amplitude) of the
changes to follow variations in the data. Figure below modulating signal. The peak amplitude and frequency
shows how amplitude modulation works. The of the carrier signal remain constant, but as the amplitude
modulating signal becomes an envelope to the carrier. of the information signal changes, the phase of the carrier
18 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

changes proportion-ately. Modulated signal is similar An AM signal consists of a carrier which acts as
to those of frequency modulation. the reference. Any modulation that is applied then
Advantages and Disadvantages of each appears as sidebands which stretch out either side of
AM the signal–each sideband is a mirror image of the other.
Advantage:
Easy to implement.
Disadvantages:
Affected by the noise signal that may add up
with the information signal. Electrical noise
causes this distortion.
As the strength of the signal decreases in a
channel with distance travelled, it reaches a
minimum level unacceptable for adequate
communication.
Within the overall AM signal the carrier possess
FM
Advantage: the majority of the power–a fully modulated, i.e. 100%
Frequency modulated wave is least affected by modulation–AM signal has sidebands which have 25%
the noise due to electrical disturbance. that of the main carrier.
Disadvantages: When demodulating a signal, two basic steps may
Frequency signal has a wide spectrum of be considered:
frequencies and therefore, need much higher Create Baseband signal: The main element
bandwidth than amplitude modulation.
of AM demodulation is to create the baseband
The no. of FM signals one can transmit over a
channel with a fixed total bandwidth is smaller signal. This can be achieved in a number of
than the no. of AM signals one can transmit ways–one of the easiest is to use a simple diode
through the same medium. and rectify the signal. This leaves elements of
PM the original RF signal. When other forms of
Advantage: demodulation are sued, they too leave some
Phase modulation offest signal modulation that elements of an RF signal.
allows computers to communicate at higher data Filter: The filtering removes any unwanted
rates through telephone system.
high frequency elements from the demo-
Disadvantage:
Phase modulation requires two signals with a dulation process. The audio can then be
phase difference between them. A reference presented to further stages for audio
pattern and a signal pattern are both necessary. amplification, etc.
AM DEMODULATION ONE TECHNIQUE ONLY The AM demodulation process is outlined in the
In order to look at the amplitude demodulation diagram below. This particular example applies
process it is necessary to first look at the format of an particularly to a diode detector.
AM signal.

Radio frequency Rectified signal Demodulated signal

signal
Basics of AM demodulation / detection
 MODULATION AND ENCODING / 19

Types of AM Demodulator broadcast applications where performance is not

There are a number of ways in which an AM signal normally an issue.
can be demodulated. There is a balance that needs to be One of the advantages of amplitude modulation
made of the performance of the circuit that is required (AM) is that it is cheap and easy to build a demodulator
against the complexity, and hence the cost that can be circuit for a radio receiver. The simplicity AM radio
tolerated. receivers AM is one of the reasons why AM has
The major types of AM demodulator are: remained in service for broadcasting for so long. One
Diode Detector: This is by far the simplest of the key factors of this is the simplicity of the receiver
form of AM demodulator or detector, requiring AM demodulator.
just a semiconductor (or other form) of diode A number of methods can be used to demodulate
along with a capacitor to remove the high AM, but the simplest is a diode detector. It operates by
frequency components. It suffers from a detecting the envelope of the incoming signal. It
number of disadvantages, but its performance achieves this by simply rectifying the signal. Current is
allowed to flow through the diode in only one direction,
is more than adequate for most applications
giving either the positive or negative half of the envelope
including broadcast receivers where cost is a
at the output. If the detector is to be used only for
significant driver.
detection it does not matter which half of the envelope
Synchronous Detector: This form of detector
is used, either will work equally well. Only when the
offers a higher level of performance, but at the
detector is also used to supply the automatic gain control
cost of considerably the use of considerably (AGC) circuitry will the polarity of the diode matter.
more components. This means that it is only The AM detector or demodulator includes a
used in receivers where the levels of capacitor at the output. Its purpose is to remove any
performance are paramount and can justify the radio frequency components of the signal at the output.
additional component costs. The value is chosen so that it does not affect the audio
Both types of detector are widely used, although base-band signal. There is also a leakage path to enable
the diode detector is far more common in view of its the capacitor to discharge, but this may be provided by
simplicity and the fact that it is quite adequate for the circuit into which the demodulator is connected.

Rectified signal
Radio frequency The capacitor removes
signal the radio frequency element

A simple diode detector (demodulator) for AM signals

This type of detector or demodulator is called a Additionally, these detectors are susceptible to the
linear envelope detector because the output is effects of selective fading experienced on short wave
proportional to the input envelope. Unfortunately the broadcast transmissions. Here, the ionospheric
diodes used can introduce appreciable levels of propagation may be such that certain small bands of the
harmonic distortion unless modulation levels are kept signal are removed. Under normal circumstances signals
low. As a result these detectors can never provide a signal received via the ionosphere reach the receiver via a
suitable for high quality applications. number of different paths. The overall signal is a
20 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

combination of the signals received via each path and Pulse Amplitude Modulation: PCM modifies
as a result they will combine with each other, sometimes the pulses created by (PAM) to create a digital
constructively to increase the overall signal level and signal. PCM first quantizes the PAM pulses.
sometimes destructively to reduce it. It is found that Quantization is a method of assigning integral
when the path lengths are considerably different this values in a specific range to sampled
combination process can mean that small portions of instances. A quantization result is shown in
the signal are reduced in strength. An AM signal consists Figure (a).
of a carrier with two sidebands. Amplitude
If the section of the signal that is removed falls in
one of the sidebands, it will change the tone of the
received signal. However, if carrier is removed or even
reduced in strength, the signal will appear to be over Time
modulated, and severe distortion will result. This is a
comparatively common occurrence on the short waves,
and means that diode detectors are not suitable for high
quality reception. Synchronous demodulation
(detection) is far superior. (a) Quantized PAM Signal
ANALOG TO DIGITAL [DIGITIZATION]
Pulse Code Modulation (PCM) A simple method is used to assign sign and
Pulse Code Modulation (PCM) is a digitizing magnitude to values to quantized samples. Each value
process in which an analog or continuous signal is is translated into its seven bit binary equivalent. The
represented in digital or discrete form. eighth bit shows the sign.
The binary digits are then transformed into a digital signal using one of the digital-to-digital encoding techniques.
Figure (b) shows the result of the pulse code modulation of the original signal encoded finally into a unipolar
signal. Only the first three sampled values are shown.

Direction of transfer
(b) PCM

PCM is actually made up of four separate processes—PAM, quantization, binary encoding, and digital-to-
digital encoding. Figure (c) shows the entire process in graphic form. PCM is a popular in voice digitisation in
telecom systems.

PAM Quantization

Direction of transfer
–127

0001100000100110 Binary 000

encoding
+127

+ 038 +024
Digital/digital 00110010000011000
encoding
Direction of transfer

(c) From analog signal to PCM digital code

 MODULATION AND ENCODING / 21

Pulse Amplitude Modulation help PAM is not useful to data communications,

in Data Communication because although it translates the original waveform
Modulation: Modulation is the method of to a series of pulses, these pulses are still of any
mixing intelligent signal on to the carrier signal
amplitude (still an analog signal, not digital). To make
so that a weak intelligent signal can be
transmitted for long distance over a them digital, they are modified by using pulse code
transmission media. modulation (PCM).
Pulse Amplitude Modulation (PAM): The Pulse Code Modulation (PCM)
first step in analog-to-digital encoding is called The bandwidth of an AM signal is equal to twice
pulse amplitude modulation (PAM). This the bandwidth of the modulating signal and covers a
technique takes analog information, samples range centred around the carrier frequency.The shaded
it, and generates a series of pulses based on
the results of the sampling. The term sampling portion of the graph is the frequency spectrum of the
means measuring the amplitude of the signal signal.
at equal intervals. BWm = Bandwidth of the modulating signal (audio)
PAM is the base of analog-to-digital encoding BWt = Total bandwidth (radio)
called pulse code modulation (PCM). fc = Frequency of the carrier
In PAM, the original signal is sampled at equal
intervals as shown in Figure (b). PAM uses sample and Amplitude

Amplitude Amplitude

Frequency
BWm BWm
Time Time

BWt =2 X BWm
AM BANDWIDTH

(a) Analog signal (b) PAM signal SAMPLING QUANTIZATION

Using PAM and PCM, we can reproduce the wave
hold technique. At a given moment the signal level is
read, then held briefly. The sampled value occurs only from exactly by taken infinite samples, or we can
instaneously in the actual waveform, but is generalized reproduce the barest generalization of its direction of
over a still short, but measurable period in the PAM change by taking three samples.
result.

From Analog Signal to PCM Digital Code

 22 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

According to the Nyquist theorem, to ensure the The process of digitizing an analog signal starts by
accurate reproduction of an original analog signal using dividing the original signal into uniformly spaced
PAM, the sampling rate must be at least twice the highest samples as shown in Figure (b). The amplitudes of the
frequency of the original signal. So if we want to sample sample pulses rise and fall with the amplitude of the
telephone voice with maximum frequency 400 Hz. We original signal. The original signal is separated into
need a samplling rate of 8000 samples per second. individual pulses or samples each sample having a
DIGITAL TO ANALOG different amplitude based on the amplitude of the
Amplitude, Frequency and Phase Shift Keyings:
original signal. At the receiving end, these samples are
Phase shift keying or PSK uses a transition or shift from
used to reconstruct the original signal. The more
one phase to another to encode data. The presence or
frequently the samples are taken, the more accurate is
absence of a transition can be used to encode data. Figure
(a) shows an example of PSK in which 1 is represented the reconstructed waveform. The minimum number of
by the presence of transition (in this case, a 180° phase samples to replicate the original waveform is given by
shift), and 0 is represented by no phase shift. This is the Nyquist theorem. According to this theorem, for a given
case of Binary Phase Shift Keying (BPSK). signal, FS, the minimum sampling rate (Nyquist
Amplitude Shift Keying and Fold Over Error: Sampling Rate SR) to assure accurate recovery of the
Data
signal at the receiving end is twice the frequency of the
0 1 1 1 1
highest sine wave element sin (2fs) of the original
signal, or SR = 2 [sin (2fs)]
Voltage

TIME
PHASE SHIFT KEYING 1/Sr Samples
Amplitude can also be used to represent digital data. A
digital data stream is fed to a Digital to Analog Converter
(DAC) to produce an output voltage that varies with (a) Sampled
the input digital information. A group of bits from the
data stream are collected and presented to the DAC. A
different analog voltage level results from each of the
sixteen combinations of 1s and 0s of the four bits. In
this way, each voltage level is a symbol representing (b) Reconstructed
four data bits. This form of digital coding is called
Amplitude Shift Keying (ASK) modulation.
Another method for sending analog information on
digital carriers involves translating voice or other analog
signals into a train of pulses that are in turn coded into
(c) Reconstructed Waveform Using Twice
Digital format. The digital data are then transmitted and
the process reversed at the received end. This process Frequency Shift Keying: The message sent by a
combines sampling with Analog-to-Digital conversion modem has not only digital computer data converted to
(ADC) and Digital to Analog conversion. (DAC). It analog form, but also control information that helps the
reduces the noise effect inherent in analog transmission receiver sample the transmission at correct intervals.
(Fig. (a)). With the method of FSK, the modem varies the carrier
wave’s frequency. Each individual change is known as
baud. This method limits transmission to 1,800 bits per
second. Higher than this bit rate would require a wider
bandwidth than that supplied by the telephone voice
Analog In
Digital
channel.
Box Figure shows the FSK system. The two pairs of
A-D Converter
frequencies are characteristics of modems that transmit
ANALOG TO DIGITAL CONVERTER at 300 bps, using frequency shift keying in full duplex
 MODULATION AND ENCODING / 23

mode (sending and receiving at the same time). Operating FSK (Frequency Shift Keying): In FSK, the
within the 4,000 Hz allocated for the telephone voice two binary numbers are represented by two
channel, the modem that originates the session transmits different frequencies near the carrier
data by generating a carrier wave at either 1070 Hz (for frequency. FSK is less susceptible to error than
ASK.
Voice Channel PSK (Phase Shift Keying): In PSK, phase of
4,000 the signal is shifted to represent data. A binary
3,400 0 is represented by sending a singal burst of
2,225 Hz
the same phase as the previous one and binary
2,025 Hz
1,270 Hz
one is represented by sending signal burst of
opposite phase to the preceding one. This is
250 Hz
0 Hz the form of Differential Phase Shift Keying.
1 1 1 0 0 0 1 1 0 1 DIGITAL MODULATION
FSK System in Full Duplex Mode [ASK, FSK, PSK, QPSK]
The digital-to-digital encoding is the process of
0s) or 1270 Hz (for 1s). Its counterpart transmits 0s at converting binary data, a sequence of bits, to a digital
2025 Hz and 1s at 2225 Hz. With FSK, only one bit is signal. For example, data, text, numbers, graphical
encoded per frequency shift. images, audio, and video that stored in computer
Concept of QPSK of Mudulation: In phase shift memory are all sequence of bits. Line coding converts
keying (PSK) the phase of the signal is changed to a sequence of bits to a digital signal. The figure shows
represent data. A ‘off’ situation is shown by sending a the concept of line encoding.
signal burst same as the previous one. A ‘on’ situation is Line
shown by sending a signal burst of opposite phase to 01011101
Coding
the preceding phase. Thus, PSK bandwdith makes
Characteristics of Line Coding: The line coding
efficient use of the channel. When signal is to be sent at scheme can be characterized into: signal versus data
a higher bit rate, then a more efficient method using level, pulse rate versus bit rate, dc components, and a
QPSK can be used. self-synchronizations.
We can divide the line coding scheme into three
+
00 cosin 10 categories: Unipolar, Polar, and Bipolar.
Unipolar: Unipolar encoding is very simple and
very primitive. It is one of the oldest method for
–1 1 encoding. Digital transmissions systems work by
– sin + sin sending voltage pulses along a medium link with a wire
or cable. In many types of encoding, one voltage level
stands for binary 0, and another level stands for binary
01
1. The polarity of pulses refers to whether it is positive
– cosin–1 11 or negative. Unipolar is so named because it uses only
QPSK Phasor diagram
one polarity. This polarity is assigned to one of the two
In QPSK (Quadrature Phase Shift Keying), instead binary states which is 1. The other state is 0, represented
of allocating a phase shift of 180º, phase shift of by zero voltage.
multiples of 90 degrees is used. This causes each signal Unipolar encoding uses only one voltage levels.
element to represent two bits rather than one. The unipolar encoding scheme is expensive to
ASK (Amplitude Shift Keying): In the ASK, implement. The unipolar encoding scheme has at least
the two binary values are represented by two problems that makes it undesirable: a dc component
two different amplitudes of the carrier and a lack of synchronization. The average amplitude
frequency. One digit is represented by the of encoded signal is zero. This creates a dc component.
presence and one digit is represented by the Lack of synchronization is also an issue in unipolar
absence of the carrier frequency. encoding scheme. If the data contains a long sequence
24 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

of 0s or 1s, there is no change in signal during this duration that can alert receiver to potential synchronization
problems. The figure given below shows the idea of unipolar encoding scheme.

Amplitude

0 1 0 0 1 1 1 0

Time

Polar allows the receiver to synchronize its timer to the actual

Polar encoding uses two voltage levels: one positive arrival of the transmission. A string of 0s still cause the
and one negative. By using two levels, in this method problems, but they are not as likely as 1s.
the average voltage level on the line is reduced and the In NRZ-I, the signal is inverted when 1 is
dc component is removed. encountered.
Polar encoding uses two voltage levels (positive The figure below shows the NRZ-L and NRZ-I
and negative). representations of the same series of bits. In the NRZ-L
The various encoding scheme under this category sequence, positive and negative voltages have specific
are: non-return to zero (NRZ), return to zero (RZ), meanings: positive for 0 and negative for 1. In NRZ-I
sequence, the voltages are meaningless. Instead, the
Manchester and differential Manchester.
receiver looks for changes from one level to another as
Types of Polar encoding scheme are: its basis for recognitions of 1s.
Amplitude
Polar

NRZ RZ Manchester

Nonreturn to Zero (NRZ): In NRZ encoding, the NRZ-L

Time
value of signal is always either positive or negative.
There are two types of NRZ.
In NRZ-L (NRZ-Level) encoding, the level of the
NRZ-I
signal depends on the type of bit that it represents. A Time
positive voltage means the bit is at 0, where a negative
voltage means the bit is at 1; thus the level of the signal
dependent upon the state of the bit. A problem can arise
Transition because
next bit is 1.
when data contain a long stream of 0s or 1s. The receiver
receives a continous voltage and determines how many Return to Zero (RZ): The original data contains
bits are sent by relying on its clock, which may or may a sequence of 1s and 0s, a solution to the problem is to
not be synchronized with the sender clock. include somehow synchronization in the encoded signal.
In NRZ-L the level of the signal is dependent upon To ensure synchronization, there must be signal change
the state of the bit. for each bit. The receiver can use these changes to build
In NRZ-I (NRZ-Invert), an inversions of the up, update and synchronize its clock. As we have seen
voltage level represents a 1 bit. It is the transition the method of NRZ-I which accomplishes this by
between a positive and negative voltage, not the voltage sequences of 1s. But to change with every bit, we need
itself, that represents a 1 bit. A 0 bit is represented by more than two values. One solution to this problem is
no change. The NRZ-I is superior to NRZ- L due to the return to zero (RZ) encoding which uses three values:
synchronization provided by the signal each time a 1 positive, negative, and zero. In RZ, the signal changes
bit is encountered. The existence of 1s in the data stream not between bits but during each bit.
 MODULATION AND ENCODING / 25

Value Amplitude

0 1 0 1 1 1 1 0

Time

Time
In Differential Manchester encoding, the transition
at the middle of the bit is used only for synchronization.
The bit representation is defined by the inversion or
non-inversion at the beginning of the bit.
These transitions can be used Bipolar Encoding: Bipolar encoding uses three
for synchronization. voltage levels: positive, negative, and zero. The zero
level in bipolar encoding is used to represent binary 0.
The main disadvantage of RZ encoding is that it The 1s are represented by alternating positive and
requires two signal changes to encode 1 bit and, negative voltages. If the first 1 bit is represented by the
positive amplitude, the second will be represented by
therefore, occupies more bandwidth. But this is one of
negative amplitude, the third by the positive amplitude,
the most effective methods for encoding. and so on. The alteration occurs even when the 1 bits
Manchester Encoding: Manchester encoding uses are not consecutive.
an inversion at the middle of each bit for both In bipolar encoding, we use three levels: positive,
synchronization and bit representation. A negative-to- negative and zero.
A common bipolar encoding scheme is called
positive transition represents binary 1, and a positive- bipolar Alternate Mark Inversion (AMI). AMI means
to-negative transition represents binary 0. By using a alternate 1 inversion. A neutral zero voltage represents
single transition for a dual purpose, Manchester binary 0. Binary 1s are represented by alternating
encoding achieves the same level of synchronization as positive and negative voltages.
RZ but with only two levels of amplitude. The figure A modification of bipolar AMI has been developed
to solve synchronizing sequential 0s for long distance
below shows the Manchester encoding. transmission. It is called BnZS (bipolar n-zero
Amplitude
substitution). In this scheme, whenever n consecutive
Zero is One is zero occur in the sequence, some of the bits in these n
0 1 0 0 1 1 1 0 bits become positive or negative which helps in
synchronization. This substitution violates the rules of
AMI in a specified manner such that receiver knows
that these bits are actually 0s and not 1s.
Time
Amplitude

In Manchester encoding, the transition at the middle 0 1 0 0 1 1 1 0

of the bit is used for both synchronization and bit
representation.
Differential Manchester: In Differential
Manchester encoding, the inversion at the middle of the Time
bit interval is used for synchronization, but the presence
or absence of an additional transition at the beginning
of the interval is used to identify the bit. Differential
Manchester encoding requires two signals changes to
There are some other line coding schemes created
represent binary 0 but only one to represent binary 1.
for special purpose in data communications. They are:
The figure below shows the Differential Manchester 2B1Q (Two binary and 1 quarternary), MLT-3, and
encoding. HDB3.
26 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

1 1 103  103
Q. 1. What is channel? and What is Bandwidth? F=  5 sec  Hz 10–2 KHz
T 10  10 –6
105
Ans. A channel is a range of frequencies (or
equivalently wavelength) assigned for data communi-
cation. The amount of data that can be transmitted via a Q. 5. A signal has a bandwidth of 20Hz. The
given communications channel in a given unit of time. highest frequency is 60 Hz. What is the lowest
It exactly shows that how much stuff that you can send frequency?
through a connection. Ans. fh (highest frequency) = 60 kz.
f (lowest frequency, ?
B (Bandwidth) = 20 Hz.
B=f
f = 40 Hz Answer

Q. 2. What is baud rate? Explain the terms bit

rate and bit interval?
Ans. The number of signal elements transmitted
per second. A signal element consists of one or more
bits.
Bit Interval: The bit interval is the time required
to send one single bit in a digital communication.
Q. 6. If a bit rate of a signal is 100 bps, how
many bits can be sent in 5 s? How many bits in 15 s?
How many bits in 100 ms?
Sol.
bit rate = 100 bps
1 seconds = 100
5 second = 100× 5 = 500 bits
15 seconds = 100×15 = 1500 bits
Bit Rate: The bit rate is the number of bit intervals 100 m second = 100 × 10–3 ×100
per second. This means that the bit rate is the number of
bits sent in is, bits per second (bps). 1
Q. 3. What is periodic and aperiodic signals? = 100 × 100 = 10 bits.
100
Ans. A periodic signal completes a pattern with in
Q. 7. Compare and contrast Rz and AMI.
a measurable time frame, called a period and repeats
the pattern over subsequent identical periods. The Ans. Rz (Return to zero) encoding, which uses three
completion of one full pattern is called a cycle. values positive, negative and zero in Rz, the signal
An aperiodic signal changes without exhibiting a changes not between bits but during each bit, a positive
pattern or cycle that repeats over time. In data voltage means I and a negative voltage means 0.
communication however, we commonly use periodic
analog signals and aperiodic digital signals to send data
from one point to another.
Q. 4. What is frequency spectrum? Express a
period of 100 ms in micro seconds and express the
corresponding frequency in kilohertz.
Ans. The distribution of signal amplitude as a
function of frequency.
100  10 –3  10 –3 AMI (Alternate Mark Inversion) means alternate 1
T = 100 ms 100× 10–3 s  inversion. A neutral, zero voltage represents binary 0.
10 –3 Binary Is are represented by alternating positive and
 105 ×106 sec  105 us negative voltages.
 MODULATION AND ENCODING / 27

4. Baud rate is half the bit rate therefore more

effective utilization of the available bandwidth of the
transmission channel.
Q. 12. Given a bandwidth of 5000 Hz for an ASK
signal, what are the baud rate and bit rate?
Ans. In ASK the Baud rate is same as the bandwidth
which means the baud rate is 5000. But because the
Baud rate and bit rate are also the same for ASK, the bit
rate is 5000 bps.
Q. 8. Name the method which is used to convert Q. 13. How is baud rate related to transmission
the digital data into digital signal. bandwidth in ASK?
Ans. Line coding is the process of converting binary Ans. The Bandwidth requirement for ASK are
data a sequence of bits, to a digital signal e.g. data text calculated using the formula:
number, graphical, images, audio and video that are 8W = (1–d) X N band
stored in computer memory are all sequences of bits. Where 8W is the Bandwidth
Line coding converts a sequence of bits to a digital N Baud is the Baud rate
signal. d is the factor related to modulation process.
So, the minimum bandwidth required for
transmission is equal to the baud rate.
Q. 14. Find the minimum bandwidth for an FSK
signal transmitting at 2000 bps. Transmission is in
half-duplex mode, and carrier is separated by 3000
Hz.
Ans. If fc1 and fc0 are carrier frequencies
BW = Baud rate + fc1 –fc0
BW = bit rate + fc1 –fc0
Q. 9. What are advantages and disadvantages = 2000 + 3000 = 5000 Hz.
of parallel transmission? Q. 15. Find the maximum bit rates for an FSK
Ans. Advantages: Advantage of parallel signal if the bandwidth of the medium is 12,000 and
transmission is speed. Because there are dedicated wire the difference between the two carriers is 200 Hz.
paths which are used to send the bits simultaneously. Transmission is in full-duplex mode.
Disadvantages: Disadvantage of parallel trans- Ans. Because the transmission is full duplex, only
mission is costly because transmission requires n 6000 Hz, is allocated for each direction.
communication lines just to transmit the data stream, for FSK, if fc1 and fc0 are the carrier frequency.
so cost of implementation is high. BW = Baud rate + fc1 – fc0
Q. 10. The bit rate of signal is 1000. If each signal Band rate = BW – (Fc1 –Fc0)
unit carries 6 bits, what is the baud rate? = 6000 – 2000
Ans. = 4000
bit rate But because the band rate is the same as the bit
Baud Rate = number of bits per signal unit rate, the bit rate is 4000 bps.
Q. 16. Can the bit rate be less than the pulse
rate? Why or why not?
3000 Ans. Pulse rate defines the number of pulses per
= = 500 Baud/s.
6 second. A pulse is the minimum amount of time required
Q. 11. What are the advantages of QAM over to transmit a symbol the bit rate defines the number of
ASK and P5K? bits per second. If a pulse carries only I bit the pulse
Ans. 1. Noise immunity of QAM is very high. rate and the bit rate are the same.
2. QAM is best suitable for high bit rates. If the pulse carries more than I bit then the bit rate
3. Low error probability. is greater than the pulse rate.
Q. 17. Compare the difference between ASK, FSK, PSK.
Ans.
Parameter ASK FSK PSK
1. Variable Amplitude Frequency Phase
Characteristics
2. Bandwidth (Hz) 2R |f1–f0|+ + (1+ r) R (1+ r) R
3. Noise Immunity Low High High
28 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

4. Error Probability High Low Low

5. Perfomance in
Presence of Noise Poor Better than ASK Better than FSK
6. Complexity Simple Moderately Vey
7. Bit rate Suitable upto Suitable upto Suitable
100 bit/ sec about 1200 for high
bit/ sec. bit rates.
8. Detection method Envelope Envelope Coherent
Q. 18 What is the difference between AM and ASK?
Ans.
Parameter AM ASK
1. Variable characteristics Amplitude Amplitude
2. Nature of Modulating Modulating signal is analog. Modulating signal is digital
Signal

3. Modulated signal shape

4. Modulated in the carries Continuous variation in Carrier ON or OFF

amplitude accordance with the amplitude of depending on whether a 1
modulating signal or 0 is to be transmitted
5. Number of side bands Two Two
produced
6. Bandwidth 2fm (1 + r) R
7. Application Radio broadcasting. Data transmission at low
rate
8. Detection method Envelope Envelope
9. Noise immunity Poor Poor.

Q. 19. What is the difference between PSK and QPSK,

Ans.
Parameter BPSK (PSK) rate
1. Variable Characteristics of Phase QPSK
the carrier Phase
2. Type of modulation Two level (binary) A
binary bit is Four level
represented by one
phase state
3. Modulated signal shape A binary ibt is
represented by one A group of two binary bits is
phase state represented by one phase state
4. Bit rate/ Baud rate Coherent Two
5. Detection method Coherent Coherent
6. Complexity Complex Coherent
7. Application Suitable for applica- Very complex
tions that need high bit Suitable for applications
needing very high bit rates.
 MODULATION AND ENCODING / 29

Q. 20. Can we convert an analog data into digital Fig. shows a simple method of assigning sign and
signal? How? magnitude to quantized samples.
Ans. Sampling is the process by which the analog Each value is translated into its 7-bit binary
signal is sampled, we can store the binary data in the equivalent. The eighth bit indicates the sign.
computer or use line coding to further change the signal + 024 00011000 – 151 10001111 + 125 01111101
to a digital one so it can be transmitted digitally. + 038 00100110 – 080 11010000 + 110 01101110
Pulse Amplitude Modulation (PAM): An analog + 048 00110000 – 050 10110010 + 090 01011010
to digital conversion method is called pulse amplitude
+ 039 00100111 – 052 00110110 + 088 01011000
modulation (PAM). This technique takes an analog
signal, samples it and generates a series of pulses based + 026 00011010 – 127 011111111 + 077 01001101
on the results of the sampling. The term sampling means The binary digits are then transformed to a digital
measuring the amplitude of the signal at equal intervals. signal by using one of the line coding techniques. Fig
In PAM, the original signal is sampled at equal intervals. shows the result of the pulse code modulation of the
PAM uses a technique called sample and hold. original signal encoded finally into a unipolar signal.
Only the first three sampled values are shown.
PCM is made up of four separate processes-3
- PAM
- Quantization
- Binary encoding
- Line coding.

Pulse code Modulation PCM modifies the pulses Q. 21. What is the way to convert digital data
created by PAM to create a completely digital signal. into analog signal?
I'CM first quantizes the PAM pulses. Quantization is a Ans. Converting binary data or a low pariety analog
method of assigning integral values in a specific range signal to a band-pass analog signal is traditionally called
to sampled instances. modulation.
Modulation of binary data or digital-to-analog
modulation is the process of changing one of the
characteristics of an analog signal based on the
information in digital signal (0’s and l’s). When you
transmit data from one computer to another across a
public access phone line for example, the original data
are digital, but because telephone wires carry analog
signals; the data must be converted.
When we vary any one of these characteristics, we
create a different version of that wave. If we then say
that the original wave represents binary 1, the variation
can represents binary 0 or vice versa.
According to this there are three methods for
modulating digital data into an analog signal.
30 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

(ASK) Amplitude shift keying Phase shift keying (PSK) in phase shift keying, the
(FSK) Frequency shift keying phase of the carrier is varied to represent binary 1 or 0.
(PSK) Phase shift keying. Both peak amplitude and frequency remain constant as
There is a fourth mechanism that consists changes the phase changes. If we start phase of 0° to represent
in both amplitude and phase is called quadrature binary 0, then we can change the phase to 180° to send
amplitude modulation (QAM). binary 1. The phase of the signal during each bit duration
Amplitude shift keying (ASK) In amplitude shift is constant and its value depends on the bit (0 or 1).
keying the strength of the carrier signal is varied to
represent binary 1 or 0. Both frequency and phase remain
constant while the amplitude changes which voltage
represents I and which represents 0 are left to the system
designers. A bit duration is the period of time that defines
I bit. The peak amplitude during each bit duration is
constant and its value depends on the bit (0 or 1)

Quadrature Amplitude Modulation (QAM):

Quadrature amplitude modulation is a combination of
ASK and P so that a maximum contrast between each
signal unit (bit, dibit, tribit and so on) is achieved.

ASK transmission is highly susceptible to noise

interference. The terms noise to unintentional voltages
introduced on to a line by various phenomena such as
heat or electromagnetic induction created by other The number of amplitude shifts is fewer than the
sources. number of the phase shifts. Because amplitude changes
Frequency Shift Keying (FSK): In frequency shift
are susceptible to noise and require greater shift
keying, the frequency of the carrier signal is varied to
difference than do phase changes, the number of phase
represent binary 1 or 0. The frequency of the signal
during each bit duration is constant and its value depends changes used by QAM system is always large the
on the bit ( 0 or 1). number of amplitude shifts.
Q. 22. What is the method to change the analog
data into analog signal?
Ans. Modulation of an analog signal or analog-to-
analog conversion is the representation of analog
information by an analog signal.
Analog-to-analog modulation can be accomplished
in three ways – Amplitude Modulation (AM): In AM
transmission, the carrier signal is modulated, so that its
amplitude varies with the changing amplitudes of the
modulating signal. The frequency and phase of the
carrier remain the same only the amplitude changes to
follow variations in the information.
 MODULATION AND ENCODING / 31

and phase of the carrier signal remain constant, but as

the amplitude of the information signal changes, the
frequency of the carrier changes correspondingly.

Frequency Modulation (FM): In FM

transmission, the frequency of the carrier signal is
modulated to follow the changing voltage level
amplitude of the modulating signal. The peak amplitude
Phase Modulation: PM is used in some systems Q. 23. What is Null modem?
as an alternative to frequency modulation in PM Ans. A short RS-232 C cable that connects two
transmission, the phase of the carrier signal is modulated personal computers so that they can communicate
to follow the changing voltage level of the modulating without the use of modems. The cable connects the two
signal. The peak amplitude and frequency of the carrier computers serial parts and certain lines in the cable are
signal remains constant, but as the amplitude of the crossed over so that the wires used for sending data by
information signal changes, the phase of the carrier one computer are used for receiving data by the other
changes correspondingly. computer, and vice versa.
32

Multiplexing and Switching

3
into many logical channels so that simultaneously many
independent signals may be transmitted in a line. The
Whenever the bandwidth of a medium linking two electronic device that performs this task is called a
devices is greater than the bandwidth needs of the multiplexer. The multiplexer brings together several low
devices, the link can be shared. Multiplexing is the set speed communication lines, transforms them into one
of techniques that allows the simultaneous transmission high speed together several low speed communication
of multiple signals across a single data link. In lines, transforms them into one high speed channel and
multiplexed system, n lines share the bandwidth of one reverses the operation at the other end.
link. As data and telecommunications usage increases, In many applications, several terminals are
the traffic also increases. We can accommodate this
increase by continuing to add individual lines each time connected to a computer. If each terminal is operating
a new channel is needed. In this chapter, we discuss at 1440 bits per second over a communication line that
various techniques for multiplexing such as FDM, TDM, can operate at 56,000 bits per second, then there is a
Synchronous TDM, and Statistical TDM. very inefficient operation. A channel is an expensive
Other topologies employing multipoint connections
are also used. Because the distances between devices resource. Multiplexing thus helps in its optimal
and the total number of devices increase beyond the utilization by simultaneously transmitting multiple
capacities of the media and equipment. A better solution signals over it.
to handle this situation is switching. A switched network A multiplexer puts several data communication
consists of a series of interlinked nodes, called switches. lines or signals helps into one data communication line
Switches are hardware and software devices capable of
creating temporary connections between two or more or signal at the sending location. For example, in the
devices. In switched network, some of these nodes are Figure below, there are 4 terminals connected to a
connected to the communicating devices. Three methods multiplexer. The multiplexer takes the signals from the
of switching that are important such as circuit switching. 4 terminals and converts them into 1 large signal which
Packet switching, and message switching. is transmitted over 1 communication channel. At the
receiving end, a demultiplexer takes the large signal and
CONCEPT reconverts it into the original 4 signals. Without
Need of Multiplexing in Communication: Multi- multiplexers, one needs 4 separate communication
plexing is the method of dividing a physical channel channels.

1
MULTIPLEXER

MULTIPLEXER

2
COMPUTER
MODEM

MODEM

4 A MULTIPLEXED SYSTEM
 MULTIPLEXING AND SWITCHING / 33

FREQUENCY DIVISION Radio transmission is an example of FDM. Each

MULTIPLEXING (FDM) radio station is assigned a frequency range within a
The bandwidth or range of a medium is always more bandwidth of radio frequencies. Several radio stations
than that of any given signal. This fact is used in
may be transmitting electromagnetic signals simul-
frequency division multiplexing. In FDM, the available
taneously over the physical channel. A radio receiver’s
bandwidth of a physical medium is divided into several
smaller, disjoint logical bandwidths. Each of the antenna receives signals transmitted by all the stations.
bandwidth is used as a separate communications line. The tuning dial in the radio is used to isolate the speech
Figure below shows process of FDM. signal of the station tuned.

FREQUENCY DIVISION MULTIPLEXING

In FDM, the signals to be transmitted are analog signal. Thus, digital signals are converted to analog form
FDM.
TIME DIVISION MULTIPLEXING (TDM) Individual messages are reassembled at the receiving
The bit rate of a transmission medium is always end. The process of TDM is illustrated in Figure below
more than the rate of the digital signal. This fact is for multiplexing three different signals.
utilized for time division multiplexing. In TDM, the total The idea can be extended for multiplexing several
time available in the channel is divided between several signals. TDM is used to multiplex digital or analog
users and each user of the channel is allotted a time- signals. For communications of digital data, it is more
interval during which he may transmit a message. The convenient to transmit data directly in digital form and
channel capacity is fully utilized in TDM by interleaving thus TDM is more appropriate. Moreover,
a number of data streams belonging to different users communication between computers occurs in short, fast
into one datastream. Streams of data sent through the bursts. Each burst would thus need the full channel
physical channel are demultiplexed at the receiving end. bandwidth which is available to a signal in TDM.

TIME DIVISION MULTIPLEXING

SDM STDM systems are comparatively easy to

Synchronous TDM as signs time slots of equal implement once the software allocates the time slots.
length to all packets regardless whether or not anything STATISTICAL TDM
is to be sent by each station with an assigned time slot. Statistical Time Division Multiplexing: STATDM
For example, if message A is not included, then its does not make a fixed assignment of time slots so that
allotted time would still be allocated. However, time any port which is idle does not receive a (full) slot. In
slots for message A would not contain information. order to identify which slot corresponds to which data
34 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

stream, it is necessary to append address and control Fibre in the loop (FITL) is a common method of
symbols to each slot that is used. This ‘overhead’ is multiplexing, which uses optical fibre as the backbone.
usually small and is more than compensated for by the It not only connects POTS phone lines with the rest of
increased efficiency derived from not having to take up the PSTN, but also replaces DSL by connecting directly
channel space with idle bits. to Ethernet wired into the home. Asynchronous Transfer
These systems are more complex but allow Mode is often the communications protocol used.
reassigning of time slots that are not in use. STATDM Because all the phone (and data) lines have been
networks assign time slots only when they are to be used clumped together, none of them can be accessed except
and delete them when they are idle. The total time used through a demultiplexer. Where such demultiplexers are
for a STATDM frame varies with the amount of traffic uncommon, this provides for more-secure communi-
currently being handled. STATDM systems are most cations, though the connections are not typically
suitable for these high-density, high-traffic applications. encrypted.
The continuous messages are assigned time slots and Cable TV has long carried multiplexed television
interleaved as each channel on the send side becomes channels, and late in the 20th century began offering
active and requires communications with another the same services as telephone companies. IPTV also
depends on multiplexing.
channel. If a channel does not have any traffic, its time
Video Processing: In video editing and processing
slots are deleted and reassigned to an active channel. In
systems, multiplexing refers to the process of
this way, the interconnecting media achieves a higher interleaving audio and video into one coherent MPEG
state of utilisation than with STDM systems. transport stream (time-division multiplexing).
TDM and STATDM require a modem in order to In digital video, such a transport stream is normally
interface with the voice line, but this may be built in. a feature of a container format which may include
All modem STATDMs have at least one microprocessor metadata and other information, such as subtitles. The
with programmed and programmable functions of great audio and video streams may have variable bit rate.
diversity and are called “Smart” or “Intelligent MUXs.” Software that produces such a transport stream and/or
MULTIPLEXING APPLICATIONS container is commonly called a statistical multiplexor
Telegraphy: The earliest communication or muxer. A demuxer is software that extracts or
technology using electrical wires, and therefore sharing otherwise makes available for separate processing the
an interest in the economies afforded by multiplexing, components of such a stream or container.
was the electric telegraph. Early experiments allowed Digital Broadcasting: In digital television and
two separate messages to travel in opposite directions digital radio systems, several variable bit-rate data
simultaneously, first using an electric battery at both streams are multiplexed together to a fixed bitrate
ends, then at only one end. transport stream by means of statistical multiplexing.
Emile Baudot developed a time-multiplexing This makes it possible to transfer several video and
system of multiple Hughes machines in the audio channels simultaneously over the same frequency
1870s. channel, together with various services.
In 1874, the quadruplex telegraph developed In the digital television systems, this may involve
by Thomas Edison transmitted two messages several standard definition television (SDTV)
in each direction simultaneously, for a total of programmes (particularly on DVB-T, DVB-S2, ISDB
four messages transiting the same wire at the and ATSC-C), or one HDTV, possibly with a single
same time. SDTV companion channel over one 6 to 8 MHz-wide
Several workers were investigating acoustic TV channel. The device that accomplishes this is called
telegraphy, a frequency-division multiplexing a statistical multiplexer. In several of these systems, the
multiplexing results in an MPEG transport stream. The
technique, which led to the invention of the
newer DVB standards DVB-S2 and DVB-T2 has the
telephone.
capacity to carry several HDTV channels in one
Telephony: In telephony, a customer’s telephone multiplex. Even the original DVB standards can carry
line now typically ends at the remote concentrator box more HDTV channels in a multiplex if the most
down the street, where it is multiplexed along with other advanced MPEG-4 compressions hardware is used.
telephone lines for that neighbourhood or other similar On communications satellites which carry
area. The multiplexed signal is then carried to the central broadcast television networks and radio networks, this
switching office on significantly fewer wires and for is known as multiple channel per carrier or MCPC.
much further distances than a customer’s line can Where multiplexing is not practical (such as where there
practically go. This is likewise also true for digital are different sources using a single transponder), single
subscriber lines (DSL). channel per carrier mode is used.
 MULTIPLEXING AND SWITCHING / 35

Signal multiplexing of satellite TV and radio The connection in circuit swtiching is made at a
channels is typically carried out in a central signal switching centre, which physically links the two ends
playout and uplink centre, such as ASTRA Platform and maintains an open line between them till the
Services in Germany, which provides playout, digital communication is finished.
archiving, encryption, and satellite uplinks, as well as Circuit switching is extensively used in modern
multiplexing, for hundreds of digital TV and radio communications on the dial-up telephone network and
channels. also in private communication networks.
In digital radio, both the Eureka 147 system of Circuit switching is thus a method of networking
digital audio broadcasting and the in-band on-channel in which the communicating machines use the linking
HD Radio, FMeXtra, and Digital Radio Mondiale circuit extensively even during periods of momentary
idleness.
systems can multiplex channels. This is essentially
The exclusive monopoly continues till the circuit
required with DAB-type transmissions (where a
is released.
multiplex is called an ensemble), but is entirely optional To set up a circuit switched connection, first an
with IBOC systems. available data path is located and seized. It is then
Analog Broadcasting: In FM broadcasting and dedicated to the exclusive use of communicating
other analog radio media, multiplexing is a term parties.
commonly given to the process of adding subcarriers to Circuit switching is most suited for voice
the audio signal before it enters the transmitter, where conversation. It is certainly efficient for connections that
modulationoccurs. Multiplexing in this sense is carry large amount of data relative to data that is
sometimes known as MPX, which in turn is also an old exchanged during the call.
term for stereophonic FM, seen on stereo systems since In packet switching a message is divided into many
the 1960s. segments or packets. Each packet is treated as a separate
CIRCUIT AND PACKET SWITCHING communication. These packets are then sent to each
Circuit switching is a method of establishing station in a continuous sequence. Packets from various
physical link between the sending and receiving messages may be transmitted together and packets of
communication ends. A telephone system is an example same message may be despatched over many different
of such switching. lines.

WORKSTATION

. .
. TRANSMISSION PATH .
. .
NODE NODE
. . . . . . .. .. .
1 2
. .
. .
. .
LEGEND
DATA PACKET
CALL-ACCEPT PACKET
. . . . . ACKNOWLEDGEMENT PACKET
CALL REQUEST PACKET

Each packet has a header with a packet number and, Receiving terminals accept packets on the basis of
source and destination addresses. address identification. The packet number enables them
The packets are reassembled into their original to reassemble the messages correctly.
messages when they reach their destination. The packets Packet switching, with its efficient use of lines and no
are despatched continuously till there are messages to be waiting time, is now used in many networks. It is most
sent. suitable for transaction-oriented operations.
36 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

5 4

6 7 8 3

1 2

Data Packets: Data packets are carriers of message segments alongwith sequence and routing information.
Control Packets: Control packets are short messages that ensure smooth transmission of communications.
They contain despatch requests and acceptance, data packet receipts, etc.
The control packets initiate and keep communication going.

Trailer User Data Header

Trailer Contents: Header Contents:
Error Checking Code Beginning of Message Indicator
End-of-Message Indicator Source Address
Destination Address
Description of Data to Follow
(e.g. User Info, Control Info, etc.)
Packet Sequence Number
Routing Information
Billing Information

These above categories can further categorize as

Q. 1. State the importance of multiplexing. following:
Ans. To make efficient use of high speed
telecommunications lines, some form of multiplexing
is used. Multiplexing allows several transmission
sources to share a larger transmission capacity.
A common application of multiplexing is in long-
haul communications. Trunks on long-haul networks are
high capacity fiber, coaxial or microwave links. These
links can carry large numbers of voice and data
transmission simultaneously using multiplexing.
Q. 2. What are the types of multiplexing?
Ans. Multiplexing can be classified as:
1. Analog
2. Digital
 MULTIPLEXING AND SWITCHING / 37

Q. 3.What is FDM?
Ans. Frequency-division multiplexing (FDM) is an
analog technique that can be applied when the bandwidth
of a link is greater than the combined bandwidths of the
signals to be transmitted. In FDM signal generated by
each sending device modulate different carrier
frequencies. These modulated signals are then combined
into a single composite signal that can be transported
by the link.
Q. 4. Write a short note on WDM. Fig. TDM
Ans.Wave division multiplexing (WDM) is (2) INVERSE TDM: As its name implies, Inverse
designed to use the high data rate capability of fibre Multiplexing is the opposite of multiplexing inverse.
optic cable. The optical fibre data rate is higher than Multiplexing takes the data stream from one high-speed
the data rate of metallic transmission cable. WDM is line and breaks it into portions that can be sent across
conceptually the same as FDM, except that the several lower-speed lines simultaneously with no loss
multiplexing and demuitiplexing in value optical signals in the collective data rate.
transmitted through fibre optic channels. Following fig.
gives the conceptual view of a WI)M multiplexer and

 
 
    

demultiplexer very narrow bands of light from different

sources are combined to make a wider band of light. At Fig. Multiplexing and Inverse multiplexing

the receiver, the signals are separated by the Think of an organization that wants to send data,
demultiplexer. audio and video, each of which requires a different data
One application of WDM is the SONET network rate. To send audio, it may need a 64-kbps link. To send
in which multiple optical fiber lines are multiplexed ad
data, it may need a 128-kbps link. And to send video, it
demultiplexed. A new method called DWDM (dense
may need a 1.544-mbps link. To accommodate all these
WDM), can multiplex a very large number of channels
needs, the organization has two options. It can lease a
by spacing channels closer to one another. It achieves
1.544-mbps channel from a common carrier (the
even greater efficiency.
telephone company) and use the full capacity only
Q. 5. Explain the two types of TDM imple-
mentation and how do they differ from each other. sometimes, which is not an efficient use of the facility.
Ans. There are two methods: Q. 6. What is Bit Padding and what is
(1) TDM interleaving?
(2) INVERSE TDM Ans. In bit padding, the multiplexer adds extra bits
(1) TDM: Time Division Multiplexing (TDM) is a to a device is source stream to force the speed
digital process that allows several connections to chare relationships among the various devices into integer
the high bandwidth of a link. Instead of sharing a portion multiples of each other. When speeds are not multiples
of the bandwidth as in FDM, time is shared. Each of each other, they can be made to behave as if they
Connection occupies a portion of time in the link. Figure were, by the technique called bit padding.
gives a conceptual view of 1DM. Note that the same In multiplexing side, as the switch opens in front
link is used as in FDM: here, however, the link is shown of connection, that connection has the opportunity to
sectioned by time rather than by frequency. In the figure send a unit onto path. This process is called interleaving.
portion of signals 1, 2, 3 and 4 occupy the link Q. 7. Is bit padding is technique for FDM or
sequentially. TDM? Is the framing bit used in FDM or TDM?
38 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Ans. Bit padding is a technique for TDM. TDM is simultaneously as sub-channels in one communication
the digital way of multiplexing. Framing bits are used channel, but are physically taking turns on the channel.
in TDM to make synchronization between sender and Q.11. What are the two types of switches used
receiver, in circuit switching? Explain.
Q. 8. Define the difference between switched arid Ans. The basic two types of switchines are:
leased lines. Crossbar Switch: A Crossbar Switch connects n inputs
Ans. In switched line communications, a link that to m outputs in a grid, using electronic micro-switches
is established in a switched network, such as the (transistors) at each cross point. The major limitation
international dial-up telephone system. of this design is the number of crosspoints required.
A leased line is a symmetric dedicated service (the Connecting n inputs to m outputs using a crossbar switch
same upstream and downstream bandwidth) creating a requires n * m crosspoints. For example, to connect 1000
permanent connection between your premises and the inputs to 1000 outputs requires a crossbar with
Internet. 1,000,000 crosspoints. A crossbar with this number of
Q. 9. What are switched communications crosspoints is impractical. Such a switch is also
networks? inefficient because statistics show that, in practice, fewer
Ans. In the switched communications networks data that 25 per cent of the crosspoints are in use at any given
entering the network from a station are routed to the time. The rest are idle.
destination by being switched from node to node. For
example, in the fig. data from station A intended for
station F are send to node 4. They may then be routed
via nodes 5 and 6 or nodes 7 and 6 to the destination.
This is called switched communication networks.

Fig. Crossbar switch

Multistage Switch: The solution to the limitations

of the crossbar switch is the Multistage Switch, which
combines crossbar switches in several stages. In
multistage switching, devices are linked to switches that,
in turn are linked top other switches.

Fig. Simple Switching Network

Q. 10. Define the difference between FDM and

TDM. Fig. Multistage switch
Ans. Frequency-Division Multiplexing (FDM) is The design of a multistage switch depends on the
a form of signal multiplexing where multiple baseband number of stages and the number postages required (or
signals are modulated on different frequency carrier desired) in each stage.
waves and added together to create a composite signal. Q.12. Define about the circuit switching with
Time-Division Multiplexing (TDM) is a type of example.
digital or (rarely) analog multiplexing in which two or Ans. In telecommunications, a circuit-switching
more signals or bit streams are transferred apparently network is one that establishes a fixed bandwidth circuit
 MULTIPLEXING AND SWITCHING / 39

(or channel) between nodes and terminals before the increasing complexity and digitization of public
users may communicate, as if the nodes were physically telecommunications networks. Simple hierarchical
connected with an electrical circuit. routing schemes have been replaced with more flexible
The public switched telephone network (PSTN) is and powerful no-hierarchical schemes. Elements of
the networks of the world’s public circuit- switched circuit switching The basic elements are:
telephone networks, in much the same way that the 1. Digital switch: It is the heart of modern system.
internet is the network of the world’s public lP-based The function of the digital switch is to provide a
packet-switched networks. Originally a network of transparent signal path between any pair of attached
fixed-line analog telephone systems, the PSTN is now devices, This path is transparent in that it appears to the
almost entirely digital, and now includes mobile as well attached pair of devices that there is a direct connection
as fixed telephones. between them.
The PSTN is largely governed by technical 2. Network interface: This element represents the
standards created by the ITU-T, and uses F. 163/E.164 functions and hardware needed to connect digital
addresses (more commonly known as telephone devices, such as data processing devices and digital
numbers) for addressing. telephones etc.
Q. 13. Discuss the basic phases used in the circuit 3. Control unit: It performs three general tasks –
switching networks. First it establish connects, it is generally done on the
Ans. Communication via circuit switching implies demand i.e. on request of attached device. Second, the
that there is a dedicated communication path between control unit must maintain the connection, because the
two stations. Communication via circuit switching digital switches use time division principles, this may
involves three phase: require ongoing manipulation of the switching element.
1. Circuit establishment: The first phase a circuit Thus the control unit must tear down the connection,
is established between two stations. ‘This is the basic either in response to a respectively one of the parties or
task of circuit switching that to provide the path between for its own reasons.
the two stations. The path is connected sequence of links Q.15. What do you mean by blocking and non-
between network nodes. blocking networks?
2. Data transfer: After the path establishment, Ans. It is an important characteristic of a circuit
information can now be transmitted frame one station switching device that either it’s blocking or non-
through network to another station. The data may be blocking. Blocking occurs when the network is unable
analog or digital, depending on the nature of network. to connect two stations because all possible paths
3. Circuit disconnect: After some period of data between them are already in use. A blocking network is
transfer, the connection is terminated. such in which such blocking is possible.
Q. 14. What is circuit switching explain its On the other hand, a non-blocking network permits
elements? all stations to be connected at once and grant all possible
Ans. In circuit switching, a dedicated path is connections requested as long as the called party is free.
established between two stations for communications. Q. 16. What is in channel signalling?
Switching and transmission resources within the network Ans. With inchannel signalling, the same channel
are reserved for the exclusive use of the circuit for the is used to carry control signals as it used to carry the
duration of the connection. The connection is call to which the control signals relate. It has the
transparent. Once it is established, it appears to attach advantage that no additional transmission facilities are
devices as if there were a direct connection. Circuit needed for signalling. Basically two forms of inchannel
switching is used in public telephone network and the signalling are in use:
basis for private networks built on leased lines and using Inband signalling: It uses not only the same
on site circuit switches. Circuit switching vas developed physical path as the call it server, but it also uses the
to handle voice traffic but can also handle digital data, same frequency band as the voice signals that are
although this latter use is often inefficient. carried.Out of band signalling: It takes advantage of the
Several important aspects of circuit switching fact that voice signals do not use the fats 4-kI-lz
networks have changed dramatically in the wake of bandwidth allotted to them. A separate narrow signalling
40 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

band within the 4k1 Ii is used to send control signal. to public telecommunications networks are telephones,
Without of band signals very narrow bandwidth is but the percentage of data traffic increases year by year.
available. With such limits, its difficult to accommodate 2. Subscriber line: The link between the subscriber
in timely fashion. The another disadvantages, it that there and the network also referred as subscriber loop or local
is amount of delay from the time subscriber enters an loop. The length of local loop is actually in a range from
address and the connection is established. a few kilometres to a few tens of kilometres.
Q. 17. Describe the public telecommunications 3. Exchanges: A switching center that directly
network and what is DDS?
supports subscribers is known as an end office an end
Ans. It is the type of circuit switching network. 1’his
office will support many thousands of subscribers in a
is actually a collection of national networks
interconnected to form international service. It can be localized area.
described using four generic architectures commonly. 4. Trunks: The branch between exchanges. Trunks
1. Subscribers: The device that attach to the carry multiple voice frequency circuits using either FDM
network. It is still the case that most subscriber devices or synchronous TDM.
Q. 18. Compare datagram and virtual circuits.
Ans.
Datagram Packet Virtual Packet
1. In this approach each packet is treated inpendently. 1. Packets are not treated independently.
2. Each node chooses next node on a packet path. 2. A preplanned route is established before any
packet is sent.
3. All packets do not follow the same path. 3. All packets follow the same path.
4. Route is established for each packet. 4. Route is established for entire conversation.
5. Network may be responsible for individual packets. 5. Network may be responsible for packets
suspances.
Q. 19. Differentiate circuit switching and datagram packet switching.
Ans.
Circuit Switching Datagram Packet
1. Dedicated transmission path. 1. No dedicated path.
2. Continuous transmission of data. 2. Transmission of packets.
3. Message are not stored. 3. Packets may be stored until delivered.
4. Path is establised for entire conversation. 4. Route established.
5. Call setup delay negligible transmission delay. 5. Network may be responsible for packets suspances.
6. Computerized switching nodes. 6. Small switching nodes.
7. User responsible for message loss protection. 7. Network may be responsible for individual packets.
8. Usually, no speed or code convesion. 8. Speed and code convesion.
9. Fixed bandwidth. 9. Dynamic use of band width.
10. No overhead bits after call set up. 10. Overhead bits in each packet.
Q. 20. Write down the advantages of packet 3. When traffic becomes heavy on a circuit
switching over circuit switching. switching network, some caller are blocked, on the
Ans. Advantages of packet switching are: packet switching network, packets are still accepted,
1. Line efficiency is greater, because single node hut delivery delay increases.
to node link can be dynamically shared by many packets 4. Priorities can be used. Thus it can transmit higher
over time. In otherhand in circuit switching time on a priority packet first.
node to node link is preallocated using synchronous time Q. 21. Is the bit padding technique for FDM or
division multiplexing. TDM? Is the framing in FDM or TDM?
2. A packet switching network can perform data Ans. Bit padding technique is for TDM. It is
rate conversion. possible to multiplex data from devices of different data
 MULTIPLEXING AND SWITCHING / 41

rates. For example, device A could use our time slot, to for the speed relationship among the various devices
while the faster device B could use two. The number of into integer multiples of each other.
slots in a frame and the input lines to which they are Framing is in the TDM: The data flow of each
assigned remain fixed throughout a give system, but connection is divided into Units, and the link combines
devices of different data rates may control different one unit of each connection to make a frame. The size
numbers of those slots.
of the unit can be 1 bit or several bits. For n input
When the spuds are not integer multiples of each
connection, a frame is organized into a maximum of n
other, they can be made to behave as if they were, by a
technique called bit padding. In bit padding, the time slots, each slot carrying one unit from each
multiplexer adds extra bits to a device’s source stream connection.
Q. 22. Compare the mechanism of a space division to the mechanism of a time division switch.
Ans.
Space Division Switch Time Division Switch
1. The paths in circuit are separated from each The paths are not separated from each other
other spatially. spatially.
2. Crossbar switch and multistage switches are Two popular methods used in time-division
multiplexing: Time slot interchange and TDM
used here. bus.
In time slot interchange (TSI), if ordinary time-
3. Crossbar switch connects n inputs to m outputs division switching is used, derived task is not
in a grid using electronic micro switches at each accomplished. When a device called time-slot
cross point. interchange (TSI) is inserted, ordering of slots
based on desired connection.

Fig. Crossbar switch

Fig. (a) No Switching

Multistage Switch: It combines crossbar

switches in several stages. Devices are linked
to switches that in-turn are linked to other
swithces.
Fig. (b) Switching

4. Space-Division Switching is instantaneous. It is not instantaneous.

5. Its disadvantage is the number of crosspoints The advantage of time-division swithcing it that
required to make space-division swithcing it needs no crosspoints.
acceptable in-terms of blocking.
6. In case of multistage switch, each connection In case of TSI processing, each connection
does not create delay as in case of time-division creats delys. Each time slot must be stored by
switching. RAM, then retrieved and passed on.
42 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Q. 23. Differentiate between message, circuit and packet switching.

Ans. Comparison of Message, Circuit and Packet Switching:
Sr. No. Parameter Message Switching Circuit Switching Packet Switching

Q. 24. Is there any method available that can Noisy channel: Shannon capacity
help us to compute how fast we can send data over a In practical, we cannot have a noiseless channel,
channel? the channel is always noisy. For a noisy channel, Claude
Ans. Yes, there are two theoretical formulas were Shannon introduced a formula, called the Shannon
developed to calculate the data rate. capacity, to determine the theoretical highest data rate
1. By Nyquist for noiseless channel. for a noisy channel.
2. Another by Shannon for a noisy channel. Capacity = Band width × log2 (1 + SNR)
Noiseless Channel: Nyquist Bit Rate Where, Bandwidth is the bandwidth of the channel,
For a noiseless channel, the nyquist bit rate formula SNR is the signal-to-noise ratio, and capacity is the
defines the theoretical maximum bit rate. capacity of the channel in bits per second. The signal to
Bit Rate = 2 × Bandwidth × log2L noise ratio is the statistical ratio of the power of the
Where, Bandwidth is the bandwidth of channel signal to the power of the noise. This formula defines
L is the number of signal levels used to represent characteristics of the channel, not the method of
data, and Bit Rate is the bit rate in bits per second. transmission.
 43

Communication Mediums
4
or a keyboard. It may also be an Analog signal such as
The characteristics and quality or a data a phone call or a video signal, digitized into a bit-stream
transmission are determined both by the characteristics for example, using pulse-code modulation (PCM) or
of the medium and the characteristics of the signal. In more advancedsource coding (analog-to-digital
the case of guided media, the medium itself is more conversion and data compression) schemes. This source
important in determining the limitations of trans- coding and decoding is carried out by codec equipment.
missions. SERIAL AND PARALLEL TRANSMISSION
For guided media, the bandwidth of the signal Digital data transmission can occur in two basic
produced by the transmiting antena is more important modes: serial or parallel. Data within a computer system
than the medium in determining transmission is transmitted via parallel mode on buses with the width
characteristics. of the parallel bus matched to the word size of the
computer system. Data between computer systems is
usually transmitted in bit serial mode. Consequently, it
DIGITAL DATA TRANSMISSION is necessary to make a parallel-to-serial conversion at a
Data transmission or digital communications is the computer interface when sending data from a computer
physical transfer of data (a digital bit stream) over a system into a network and a serial-to-parallel conversion
point-to-point or point-to-multipoint communication at a computer interface when receiving information from
channel. Examples of such channels are copper wires, a network. The type of transmission mode used may
optical fibres, wireless communication channels, and also depend upon distance and required data rate.
storage media. The data are represented as an Parallel Transmission: In parallel transmission,
electromagnetic signal, such as an electrical voltage, multiple bits (usually 8 bits or a byte/character) are sent
radiowave, microwave, orinfrared signal. simultaneously on different channels (wires, frequency
While analog transmission is the transfer of channels) within the same cable, or radio path, and
continuously varying analog signal, digital synchronized to a clock. Parallel devices have a wider
communications is the transfer of discrete messages. data bus than serial devices and can therefore transfer
The messages are either represented by a sequence of data in words of one or more bytes at a time. As a result,
pulses by means of a line code (baseband transmission), there is a speedup inparallel transmission bit rate over
or by a limited set of continuously varying wave forms serial transmission bit rate. However, this speedup is a
(passband transmission), using a digital modulation tradeoff versus cost since multiplewires cost more than
method. The passband modulation and corresponding a single wire, and as a parallel cable gets longer, the
demodulation (also known as detection) is carried out synchronization timing between multiple channels
by modem equipment. According to the most common becomes more sensitive to distance. The timing for
definition of digital signal, both baseband and passband parallel transmission is provided by a constant clocking
signals representing bit-streams are considered as digital signal sent over a separate wire within the parallel cable;
transmission, while an alternative definition only thus parallel transmission is considered synchronous.
considers the baseband signal as digital, and passband Serial Transmission: In serial transmission, bits
transmission of digital data as a form of digital-to-analog are sent sequentially on the same channel (wire) which
conversion. reduces costs for wire but also slows the speed of
Data transmitted may be digital messages transmission. Also, for serial transmission, some
originating from a data source, for example, a computer overhead time is needed since bits must be assembled
44 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

and sent as a unit and then disassembled at the receiver. Twisted pair cable and coaxial cable use metallic
Serial transmission can be either synchronous or (copper) conductors that accept and transport signals
asynchronous. In synchronous transmission, groups of in the form of electric current. Optical fibre is a glass or
bits are combined into frames and frames are sent plastic cable that accepts and transports signals in the
continuously with or without data to be transmitted. In form of light.
asynchronous transmission, groups of bits are sent as There is four basic types of Guided Media:
independent units with start/stop flags and no data link 1. Open Wire
synchronization, to allow for arbitrary size gaps between 2. Twisted Pair
frames. However, start/stop bits maintain physical bit 3. Coaxial Cable
level synchronization once detected. 4. Optical Fibre
Applications: Serial transmission is between two Unguided Transmission Media consists of a means
computers or from a computer to an external device for the data signals to travel but nothing to guide them
located some distance away. Parallel transmission either along a specific path. The data signals are not bound to
takes place within a computer system (on a computer a cabling media and as such are often called Unbound
bus) or to an external device located a close distance Media.
way. Radio Link Systems:
A special computer chip known as a universal This is a connection via a chain of transmitters
asynchronous receiver transmitter (UART) acts as the and receivers.
interface between the parallel transmission of the There are radio links for both analogue and
computer bus and the serial transmission of the serial digital transfer.
port. UARTs differ in performance capabilities based Analogue radio systems can be used to transfer
on the amount of on-chip memory they possess. pulse modulated signals while digital systems
Examples are purely design for digital transmission.
Examples of parallel mode transmission include Each radio link connection requires two radio
connections between a computer and a printer (parallel channels, one in each direction. The
printer port and cable). Most printers are within 6 metres transmission frequency and the receiving
or 20 feet of the transmitting computer and the slight frequency are separated by a few MHz. This
cost for extra wires is offset by the added speed gained is a very small difference, bearing in mind the
through parallel transmission of data. frequency band use.
Examples of serial mode transmission include Satellite Systems:
connections between a computer and a modem using Satellite transmission is similar in principle to
the RS-232 protocol. Although an RS-232 cable can the ordinary radio link. Instead of having all
theoretically accommodate 25 wires, all but two of these the stations earthbound, we send some up into
wires are for overhead control signaling and not data the space.
transmission; the two data wires perform simple serial Communications Satellite rotate at almost
transmission in either direction. In this case, a computer exactly the same rate as the earth rotate.
may not be close to a modem, making the cost of parallel Compared to the radio link, the satellite has a
transmission prohibitive-thus speed of transmission may considerable large range. They are used for
be considered less important than the economical both in the national network and in the
advantage of serial transmission. international network.
GUIDED AND UNGUIDED MEDIUMS There are only a few problems in the
Guided media, which are those that provide a transmission characteristics of the satellite link.
conduit from one device to another, include twisted- Due to the long distance that the signals have
pair cable, coaxial cable, and fibre-optic cable.
to travel, resulted in a delay (echo) which
Guided Transmission Media uses a “cabling”
have to be counteracted by the echo
system that guides the data signals along a specific path.
suppressors. It has to be recognized that this is
The data signals are bound by the “cabling” system.
Guided Media is also known as Bound Media. Cabling a communication between two floating bodies
is meant in a generic sense in the previous sentences in space thus there is always a relative move-
and is not meant to be interpreted as copper wire cabling ment between the earth and the satellite which
only. Cable is the medium through which information can cause errors in digital transmissions.
usually moves from one network device to another. However, this may be compensated for by
 COMMUNICATION MEDIUMS / 45

intermediate storage of the information in rotation speed. It appears to be stationary relative to

buffer memories. earth and always stays over the same point with respect
The capacity of the telephone channels to earth. This allows a ground station to aim its antenna
increase as time goes on as compared to the at a fixed point in the sky as shown in the Figure below:
time when the satellite was first launch into
Satellite in Space
space. The Intelsat which was first launch in
1965 have 75 duplex telephone channel but
today the basic version of the new Intelsat VI 6 GHz 4 GHz
satellite can handle 80 000 telephone channel.
WIRELESS COMMUNICATION
Microwave Transmission
Above 100 MHz, the waves travel in straight lines
and can therefore be narrowly focussed. Concentrating
all the energy into a small beam using a parabolic Transmitting Receiving
antenna (like the satellite TV disk) gives a much higher Station on earth Station on earth
signal to noise ratio. Before the advent of fibre optics, SATELLITE COMMUNICATION FROM
these microwaves were used media for long distance ONE POINT TO ANOTHER
telephone system.
Microwave signals at 6 GHz are transmitted from
To overcome the problems of line-of-sight and
a transmitter on earth to a satellite. By the time this signal
power amplification of weak signals, microwave
reaches the satellite it becomes weak due to distance
systems use repeaters at intervals of about 25 to 30 km
travelled (36,000 kms). The transponder in a satellite
in between the transmitting and receiving stations. The
amplifies the weak signals and sends them back to the
first repeater is placed in line-of-sight of the transmitting
earth at a frequency of 4 GHz. These signals are received
station and the last repeater is placed in line-of-sight of
at a receiving station on the earth. The transmitting
the receiving station. Two consecutive repeaters are also
frequency is different from the receiving frequency of
placed in line-of-sight of the receiving station. Two
the statellite. This avoids interference of the powerful
consecutive repeaters are also placed in line-of-sight of
re-transmitted signal with the weak incoming signal.
each other. The data signals are received, amplified, and
re-transmitted by each of these stations. The range of frequency and type of medium used
Unlike radio waves, at lower frequencies, is shown in Figure below. Radio waves can be broadcast
microwaves do not pass through buildings well. Some omni-directionally or directionally. Various kinds of
waves may be refracted off low-lying atmospheric layers antennas can be used to broadcast radio signals. The
and may take slightly longer to arrive than direct waves. power of the radio frequency (RF) signal is determined
The delayed waves arrive out of phase with the direct by the antenna and trans-receiver (a device that
wave and thus cancel the signal. This effect is called TRANSmits and reCEIVEs a signal over medium such
Multipath Fading. It is often a serious problem in as copper, radio waves, or fibre-optic cables).
microwave communication systems. In vacuum, all electromagnetic waves travel at the
Since microwaves travel in a straight line, if the same speed, no matter what their frequency is. This
towers are too far apart, the earth will get in the way. speed, usually called the speed of light, c, and it is
Consequently, repeaters are needed periodically. The approximately 3 × 108 metres per second. In copper or
higher the towers are, the further apart they can be. The fibre the speed slows to about 2/3rd of this value and
distance between repeaters goes up very roughly with becomes slight frequency dependent. The fundamental
the square root of the tower height. For 100 metre high relation between frequency, (f), wavelength l, and c (in
towers, repeaters can be spaced 80 km apart. vacuum) is
A communication satellite is basically a microwave lf = c
relay station placed precisely at 36,000 km above the For example, 1 MHz waves are about 300 metres
equator where its orbit speed exactly matches the earth’s long and 1 cm waves have a frequency of 30 GHz.
46 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

addressed, the popularity of wireless LANs has grown

rapidly.
f(Hz)100 102 104 106 108 1010 1012 1014 1016 1018 1022 1024

Wireless LANs can operate in one of two

Radio Micro Gamma
Infra- CV X-ray
wave red ray

configurations: with a base station and without a base

station. Now it is the time to take a closer look at the
Visible light

f(Hz)104 105 106 107 108 109 1010 1011 1011 1013 1016 technology.
Twisted pair Satellite Fibrer
optics A wireless LAN or WLAN or wireless local area
network is the linking of two or more computers or
Coax Terrestrial
Maritine
devices using spread-spectrum or OFDM modulation
AM FM
radio radio
TV
technology based to enable communication between
Band LF MF HF VHF UHF SHF EHT THF devices in a limited area. This gives users the mobility
RADIO FREQUENCY RANGE AND TYPE OF
TRANSMISSION MEDIA
to move around within a broad coverage area and still
be connected to the network.
Wireless LAN For the home user, wireless has become popular
As the name suggests, a wireless LAN is one that due to the ease of installation, and location freedom
makes use of a wireless transmission medium. Until with the gaining popularity of laptops. Public businesses
relatively recently, wireless LANs were little used; such as coffee shops or malls have begun to offer
the reasons for this included high prices, low data wireless access to their customers; some are even
rates, occupational safety concerns, and licensing provided as a free service. Large wireless network
req uirements. As these problems have been projects are being put up in many major cities.
COAXIAL CABLE
A coaxial cable has a stiff copper wire as the core surrounded by insulating material. The insulator is encased
by a cylindrical conductor, a closely woven braided mesh. The outer conductor is covered in a protective plastic
sheath. A sectional view of coaxial cable is shown in the Figure. The signal is transmitted by inner copper wire and
is electrically shielded by the outer metal sleeve.

Protective
Wire mesh plastic
Insulating
conductor covering
material
Copper
core

COAXIAL CABLE

Advantages and disadvantages of coaxial cables thousands of channels of voice-grade and/or

Better shielding than twisted pairs, so it can low-speed data over a single cable. CD quality
cover longer distances at higher data rates. audio (1.4 Mbps, or a digital bit stream at
Can be used for both analog and digital data 3 Mbps) can be mixed on coaxial cable for
transmission. For analog, 75 ohm, broadband transmitting video signal.
and for digital 50 ohm, baseband cable is FIBRE OPTIC CABLE
used. Fibre optic is the latest form of bounded media. It
Coaxial cable has higher bandwidth and is superior in data handling and security characteristics.
excellent noise immunity.
The fibre optic cable transmits light signals rather than
Relatively cheap as compared to fibre optic
cables and easy to handle. electrical signals. It is far more efficient than the other
Coaxial cable has bandwidth in the range of network transmission media. Each fibre has an inner
300-400 MHz. It is capable of carrying over core of glass or plastic that conducts light. There are
50 standard 6 MHz colour TV channels or two types of light sources. These sources of lights are:
 COMMUNICATION MEDIUMS / 47

Electrical
Signal
Electrical Optical fibre
Signal Electrical to Light to
light wave electrical
converter wave converter
Light waves Light waves
TRANSMISSION THROUGH OPTICAL FIBRES

Light Emitting Diodes (LEDs) Table: Comparison of Semiconductor

Light Amplification by Stimulated Emission Diodes Laser and LEDs as Light Source
Radiation (Lasers)
Item Light Emitting Semiconductor
Figure above shows the principle of operation of Diode (LED) Laser
the fibre optic system. In single mode fibres the core is
8 to 10 microns (about the size of hair). In multimode Data Rate Low High
fibres, the core is 50 microns in diametre. Mode Multimode Multimode or
At the source is a converter that transforms Single mode
electical signals into light waves. These light waves Distance 3 km 30 km
are transmitted over the fibre. Another converter Lifetime Long life Short life
placed near the sink converts the light waves back to
Temperature Minor Substantial
electrical signals by photoelectric diodes. These
electrical signals are then amplified and sent to the Sensitivity
receiver. Cost Low Substantial

Structure of Single and Multimode Optical Fibre

Each fibre has an inner core of glass or plastic that conducts light. The inner core is surrounded by cladding, a
layer of glass that reflects the light back into the core. Each fibre is surrounded by a plastic sheath. The sheath can
be either tight or loose.
Optical fibre is either multimode or single mode. Single mode fibres allow a single light path and typically use
laser signalling. Single mode fibre allows greater bandwidth and cable run than multimode, but is expensive.
Multimode fibres use multiple light paths. The physical characteristics of the multimode fibre make all parts of the
signal arrive at the same time, appearing to the receiver as though they are one pulse. following Figure shows the
single mode and multimode optical fibres.
Single mode
Jacket
(Sheath)

Fiber
Cladding Core
Core Cladding

(a) Fiber-optic cables with

tight and loose sheaths (b) Single-mode and
multi-mode optical fibres Multimode

Optical fibres are differentiated by core/cladding 50-micron core/125-micron cladding,

size and mode of operation. Micron is one millionth of multimode
a metre = 1/25,000 inch (approx.)
100-micron core/140-micron cladding,
The following are the common types of fibre-optic
cable: multimode
8.3-micron core/12.5-micron cladding, Components of a Typical Fibre Optic Link
single-mode The figure given below shows the different
62.5-micron core/125-micron cladding, components of a fibre optic system. Signal losses are
multimode due to the following reasons:
48 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Absorption and attenuation of the cable Loss of light also occurs because of bands in
because the cladding is not completely laying in the cable.
opaque. So some of the light energy is TWISTED PAIR CABLES
absorbed into the cladding. A twisted pair consists of two insulated copper wires
Large l osses result from the p hysical of about 1 mm thick. The wires are twisted together in a
connections that bring light sources and helical. The twisting reduces electrical interference from
similar pairs close by.
detectors into alignment with the fibre
Twisted pair wires [Figure (a)] are generally used
cable.
in local telephone communication, and for digital data
Losses also occur between the splices that
transmission over short distance up to 1 km. When in
connect two cables together. large number they are bundled together and put in a
Misalignment of the light source to cable cause protective sheath. The pairs in such bundles would have
loss of light energy. interfered with one-another but for the twisting.

Electrical Signal Electrical Signal

Optical fibre Light to
Electrical to
light wave electrical
converter wave converter
Light waves Light waves
COMPONENTS OF A FIBRE OPTIC LINK

Comparision of Optical Fibre Cable and Copper Wire (UTP, STP, Coaxial and Fibre Optic Cables)
Factor Unshielded Shielded Coaxial Fibre
Twisted Pair Twisted Pair Cable Optic
Cable (UTP) Cable (STP)

1. Cost Low Moderate Moderate High

2. Installation Easy Easy Easy Difficult 2 Gbps
3. Bandwidth 1 to 155 1 to 155 Mbps typically (typically 100
10 Mbps Mbps)
4. Capacity 10 Mbps (typically
16 Mbps)
5. Attenuation High High Low Lowest
6. Electromagnetic Most
Interference vulnerable Less vulnerable Less Not affected by
(EMI) than UTP vulnerable EMI or
than UTP evesdropping
but still vulnerable
vulnerable to EMI
to EMI

Twisted pairs are normally used to connect Can be used for both analog and digital data
terminals to the main computer upto short distances from transmission. Several megabits per second can
the main computer. Data transmission speeds of upto be achieved for a few kilometres.
9600 bits per second are possible for a distance upto100 Least costly for short distances.
If part of a twisted pair cable is damaged, the
metres. entire network is not shutdown, as with coaxial
Advantages of Twisted Pair Cable cable.
Manpower to repair and service are easily Disadvantages of Twisted Pair Cable
available. Easily affected by noise signals which leads to
In telephone system, signal can travel several higher error rates, when the line length is more
kilometres without amplification. than 100 metres.
Being thin in size, likely to break easily.
 COMMUNICATION MEDIUMS / 49

Can support only 19,200 bps upto 50 metre on Advantages of Twisted Pair Wire
RS-232 port. Manpower to repair and service are easily
TWISTED PAIR WIRE available.
A twisted pair consists of two insulated copper wires In telephone system, signal can travel several
of about 1 mm thick. The wires are twisted together in a kilometres without amplification.
helical. The twisting reduces electrical interference from Can be used for both analog and digital data
similar pairs close by. transmission. Several megabits per second can
Twisted pair wires [Figure (a)] are generally used be achieved for a few kilometres.
in local telephone communication, and for digital data Least costly for short distances.
transmission over short distance up to 1 km. When in If part of a twisted pair cable is damaged, the
large number they are bundled together and put in a entire network is not shutdown, as with coaxial
protective sheath. The pairs in such bundles would have cable.
Disadvantages of Twisted Pair Wire
interfered with one-another but for the twisting.
Easily affected by noise signals which leads to
Twisted pairs are normally used to connect
higher error rates, when the line length is more
terminals to the main computer upto short distances from than 100 metres.
the main computer. Data transmission speeds of upto Being thin in size, likely to break easily.
9600 bits per second are possible for a distance upto100 Can support only 19,200 bps upto 50 metre on
metres. RS-232 port.

(a) A TWISTED PAIR OF WIRES Jacket

Shield Two
twisted
pair

(b) Unshielded twisted pair (UTP) cable (c) Shielded four pair cable

Shielded wire (Figure (c) is used in an electrically 10 Mbps. It is now the standard cable for most telephone
noisy environment to limit the effects of noise systmes.
absorptions. Unshielded twisted pair, (UTP) is the more Category 4: It also valid for Category 3 cables as
common of the two configurations. Twisted pair wiring well as other conditions to bring the possible
is commonly used for LAN media. The twisted pair transmission rate to 16 Mbps.
version of Ethernet is designated as 10BASE-T, in which
Category 5: Used for data transmission upto 100
10 refers to the Ethernet clock rate of 10 Mbps.
CONNECTORS Mbps.
The Electronic Industries Association has
developed standards to grade UTP cables by quality.
Categories are determined by cable quality with 1 as Q. 1. What are the measurements available in
the lowest and 5 as the highest. Each EIA category is the data communication?
suitable for certain uses and not for others. Ans. The four measurements used in data
Category 1: The basic twisted pair cabling used communication are throughput, propagation speed,
in telephone systems. This level of quality is fine for propagation time and wavelength.
voice lent inadequate for all but low speed data
Throughput: The throughput is the measurement
communication.
Category 2: The next higher grade suitable for of how fast data can pass through an entity. In other
voice and for data transmission of upto 4 Mbps. words, it we consider this entity as a wall through which
Category 3: Required to have atleast three twists Bits pass throughput is the number of bits that can pass
per foot and can be used for data transmission of upto this wall in one second.
50 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Propagation Speed: It measures the distance that Analog

a bit or signal can travel through the medium in one 1. While analog signals are always subject to some
second. This factor depends upon electromagnetic sort of degradation in signal quality and information
signals depends on the medium and on the frequency of loss.
the signal. 2. On the other hand, analog system have high
budget to implement.
3. It is little bit typical to calibrate and adjust.
4. It does not have resistance to noise. Noise effects
more in analog communication or transmission system.
5. Analog transmission is slower than that of digital
transmission.
In conclusion, the strength of using a digital system
over analog is clear. Digital signals are easier to transmit
and offer less room for error to occur. This leads to
accurate data transmission that in turn leads to faster
transmission rates and better productivity.
Propagation Time: This is the time required for a Q. 3. How do guided differ from unguided
signal or a bit to travel from source 10 destination. The media? Name some types of guided and unguided
propagation time is calculated by dividing the distance media.
by propagation speed. Ans. Guided mediums are used for point-to-point
Propagation time = Distance/Propagation speed connections between two devices while unguided
Wavelength: Wavelength binds the period or the mediums allow signals to be sent out informally across
frequency of a simple sine wave to the propagation speed multiple areas reaching a wide variety of systems.
of the medium. While the frequency of a signal is Guided:
independent of the medium, the wavelength depends (i) Twisted pair cable
on both the frequency and the medium. This is the (ii) Co-axial cable
(iii) Fibre optic cable.
distance a simple signal can travel in one period.
Unguided (wireless):
(i) Radio wave
(ii) Microwave
(iii) Infrared.
Q. 4. What is the form of the signal in twisted
pair cable and co-axial cable? How does this differ
from the signal in fiber-optic cable?
Ans. The signal in twisted pair cable and co-axial
Q. 2. Difference between analog and digital data cable transport signals in the form of electric current.
transmission? And which is the best way of Whereas, the fiber optics cable transport signals in the
transmission? form of light.
Ans. As a technology analog is the process of taking Q. 5. What is the connector used for twisted pair
cable?
an audio or video signal (the human voice) and
Ans. The most common twisted pair cable is RJ 45
translating it into electronic pulses. Digital on the other
(RJ stands for Registered jack).
hand is breaking the signal into a binary format where Q.6.What do you understand by wireless
the audio or video data is persecuted by a series of “1 transmission? What are three ways for wireless data
S” and “0 S”. Simple enough when its S the device to be propagated?
analog or digital phone, fax, modem, or likewise does Ans. Wireless communication is the transfer of
all the converting for you. information over a distance without the use of electrical
Digital conductors or “wires”. The distances involved may be
1. Digital signal transfer is assumed to be error free. short (a few meters as in television remote control) or
2. Digital systems have cheap electronic circuits. very long (thousands or even millions of kilometres for
3. Easier to calibrate and adjust. radio communication). When the context is clear the
4. Digital communication has resistance to noise. term is often simply shortened to “wireless”. Wireless
5. Digital transmission is the faster way of communication is generally considered to be a branch
communication. of telecommunications.
(i) Radio
 COMMUNICATION MEDIUMS / 51

(ii) Micro Q. 14. Difference between error control and flow

(iii) Infrared control.
Q. 7. What is ground propagation? Ans. Error control: In data communications,
Ans. Radio waves travel through the lowest portion errors are inevitable. Using better equipments and more
of the atmosphere, hugging he earth. ‘these low reliable transmission media may reduce the severity or
frequency signals emanate in all directions from the the frequency of occurrence, but it can never eliminate
transmitting antenna and follow the curvature of the errors.
planet. Networks must be able to transfer data from one
Q. 8. What is the purpose of cladding in an device to another with complete accuracy.
optical fibre? Flow control: In most protocols, flow control is a
Ans. Optical fibres use reflection to guide light set of procedures that tells the sender how much data it
through a channel. A glass or plastic core is surrounded can transmit before it must wait for an acknowledgement
by a cladding of less dense glass or plastic. The from the receiver. The flow of data must not be allowed
difference in density of the two materials must be such to overwhelm the receiver.
that a beam of light moving through the core is reflected Q. 15. Write a short note on twisted pair and
off the cladding of being refracted into it. what are the various standards for twisted pair
Q. 9. What is the connector used to connect cable?
devices through co-axial cable? Ans. A twisted pair consists of two conductors
Ans. The most common type of connector is BNC (normally upper, each with its own plastic insulation,
(Bayone-Neill-Concelman). Connector the BNC T twisted pair).
connector is used in Ethernet networks.
Q. 10. Define Shannon capacity and its
functions?
Ans. Shannon capacity shows the relationship
between the rate of data transmission and the amount
of noise that can affect transmissions. It states that the
faster data is transmitted, the more susceptible it is to
One of the wires is used to carry signals to the
noise. It is a function of channel capacity.
receiver and the other is used only as a ground reference.
Q. 11. Why twisted pair cables are twisted
In addition to the signal sent by the sender on one of the
together?
wire, interference (noise) and crosstalk may affect both
Ans. The wires in twisted pair cabling are twisted wires and create unwanted signals. By twisting the cable
together in pairs. Each pair would consist of a wire used balance is maintained.
for the +ve data signal and a wire used for the -ve data The most common twisted pair cable used in
signal. Any noise that appears on I wire of the pair would communication is referred to as
occur on the other wire. Because the wires are opposite Unshielded twisted pair (UTP) and
polarities they are 180 degrees out of phase when the Shielded twisted pair (STP)
noise appears on both wires it cancels out at the receiving 1. STP cable has a metal foil or traided-musk
end so, they are twisted. covering that encases each pair of insulated conductors.
Q. 12. What is the connector used to connect Although metal casing improves the quality of cable by
fibre optic? preventing the penetration of noise or crosstalk it is
Ans. The subscriber channel (SC) connector is used bulkier and more expensive.
in cable TV. It uses a push/ pulls locking system. The 2. UTP Cable has a metal which is used for wireless
straight tip (ST) connector is used for connecting cable communication media. These help to improve the
to networking devices. MT-RJ is a new connector with Analog Signals.
the same size as Rj-45.
Q. 13. What is the difference between the
omnidirection and unidirectional?
Ans. Omnidirection means to propagate signal in
all directions.
Radio waves have the characteristics of
omnidirection.
Unidirectional means to propagate signal in aligned
form, it will focus only the one particular direction to
propagate the signal.
52 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Q. 16. What are various connectors used in

(i) Twisted pair cable
(ii) Co-axial cable
(iii) Fibre optics cable
Ans. (i) The most common UTP connector is RJ-
45 (Registered Jack). The RJ-45 is keyed connector,
meaning the connector can be inserted in only one way. Fibre-optics cable connectors: Fibre-optic cables
use there different types of connectors.
The subscriber-channel. (SC) connector is used in
cable TV.
It uses a push/pull locking system. The straight-tip
(ST) connector is used for connecting cable to
networking devices. It uses a bayonet locking system
and is more reliable this SC MT-RJ is a new connector
with the same size as RJ 45.

(ii) To connect co-axial cable to devices, we need

coaxial connectors. The most common type of connector
used today is the Bayone-NeiIl-Concelman, or BNC,
connectors.
Q. 17. What is difference between twisted pair, co-axial cable and fibre optics?
Ans.
Twisted pair cable Co-axial cable Optical fibre
1. Transmission of signals 1. Transmission of signals 1. Signal transmission
takes place in the electrical takes place in the electrical takes place in an optical
form over the metallic form over the inner formas over a glass
conducting wires. conductor of the cable. fibre.
2. In this medium the noise 2. Coaxial having higher 2. Optical fiber has highest
immunity is low. noise immunity than noise immunity as the
twisted pair cable. light rays are unaffected
by the electrical noise.
3. Twisted pair cable can be 3. Coaxial cable is less 3. Not affected by the
affected due to external affected due to external external magnetic field.
magnetic field. magnetic field.
4. Cheapest medium. 4. Moderate Expensive. 4. Expensive
5. Low Bandwidth 5. Moderately high band- 5. Very high bandwidth
width.
6. Attenuation is very high. 6. Attenuation is low. 6. Attenuation is very low.
7. Installation is easy. 7. Installation is fairly easy. 7. Installation is difficult.

Q. 18. What is Radio wave transmission system? (ii) Very high frequency used in FM radio and TV.
Ans. Radio wave have frequencies between 10 kHz (iii) Ultra high frequency used is TV.
and 1 gigahertz. The range of electromagnetic spectrum
The radio frequency bands are regulated and require
between 10 kHz and I GHz is called radio frequency
(RF). a regulatory body. Radio waves can broadcast
Radio Frequency includes the following types: omnidirectionally or directionally. Various kinds of
(i) Short wave used in AM radio. antennas are used to broadcast these signals in figure.
 COMMUNICATION MEDIUMS / 53

(v) Other advantages: No cross talk inside the

optical fibre cable.
(vi) Signal can be sent up to 100 times faster. Twisted pair: The wires is twisted pair cabling are
Q. 20. What is guided media? twisted together in pairs.
Or Each pair would consist of wire used for the +ve
What are guided and unguided media? data signal and a wire used for the –ve data signal. Any
Or noise that appears on +ve/–ve wire of the pair would
What are the various media types available? occur on the other wire. Because the wires are opposite
Or polarities, they are 180 degrees out of phase (180 degree
Write short on: phases or definition of opposite polarity) when the noise
- Coaxial appears on both wires, it cancels or nulls itself out at
- Optical Fibre the receiving used. Twisted pair cables are most
- Twisted pair effectively used in system that use a balanced line
- Open wire method of transmission. Polar line coding ( Manchester
Or encoding) as opposed to unipolar line coding.
Write Note on
- Radio wave
- Micro wave
- Infrared The degree of reduction in noise interference is
Ans. There are two basic categories of transmission determined specially by the number of turns per foot
media: increasing the number of turns per foot reduces the noise
1. Guided Media. interference. To further improve noise rejection, a foil
2. Unguided. or wire braid shield is wound around the twisted pair.
Guided Transmission media uses a cabling system This shield can be moves around individual pairs or
that guides the data signals along a specific path. The around a multi pair conductor.
data signals are found by the cabling system. Guided Cables with the shield are called shielded twisted
media is also known as Bound media. Cabling is meant pair and commonly abbreviated SIP. Cables without a
in a generic sense in the previous sentences and is not shield are called unshielded twisted pair or UTP.
Twisting the wires together results in characteristics
meant to be interpreted as copper wire cabling only.
impedance for the cable.
Unguided transmission media consists of a means
UTP or unshielded twisted pair cable is used on
for the data signals to travel but nothing to guide them Ethernet 10 Base T and can also be used with taken
along a specific path. The data signals are not bound to ring, It uses the RJ line of connectors (RJ 45, RJ 11
a cabling media and as such are after called unbounded etc).
media. STP or shielded twisted pair is used with the
There are four basic types of Guided media: traditional Token Ring Cabling.
(i) Open wire Coaxial Cable: Coaxial cable consists of 2
(ii) Twisted pair conductors. The inner conductor is held inside the
(iii) Coaxial cable insulator with the other conductor waves around it
(iv) Optical fibre providing a shield, An insulating protective coating
Open wire: Open wire is traditionally used to called a jacket covers the outer conductor.
describe the electrical wire strung along power roles.
There is single wire strung between poles. No shielding
or protection from noise interference. This can include
multiconductor cables or single wire. This media is
susceptible to a large degree of noise and interference
and consequently not acceptable for data transmission
except for short distances under 20 ft.
54 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

The outer shield protects the inner conductor from building. It is the disadvantage because we cannot isolate
outside electrical signals. The distance between the outer a communication to first inside or outside a building.
conductor (Shield) and inner conductor plus the type of The radio waves band is relatively narrow just under I
material used for insulating the inner conductor GHz, compared to the microwave band. When this band
determine the cable properties or impedance. Typical is divided into subband, the sidebands are also narrow,
impedance for coaxial cables is 75 ohm for cable TV, leading to a low data rate for digital communications.
50 ohm for Ethernet thermite and thicknet. The excellent
control of the impedance characteristics of the cable
allow higher data rates to be transferred than twisted
pair cable.
Optical Fibre: Optical fibre consists of thin glass
fibre that can carry information at frequencies in the
visible light spectrum. The typical optical fibre consists
of a very narrow strand of glass called the cladding. A
typical core diameter is 62.5 microns. Typically cladding
has a diameter of 125 minors. Coating the cladding is a Microwaves: Electromagnetic waves having
protective coating consisting of plastic, it is called the frequencies between I and 300 GFL are called
jacket. microwaves.
Microwaves are unidirectional, when an antenna
transmits microwaves they can be narrowly focused.
This means that the sending and receiving antennas need
to be aligned. The unidirectional property has an obvious
advantage. A pair of antennas can be aligned without
interfering with another pair of aligned antennas.
On the other hand microwaves propagation is line-
of-sight. Since the towers with the mounted antennas
needs to be in direct sight of each other, towers that are
Q. 21. What is Unguided Transmission or what for apart need to he very tall, the curvature of the earth
is Unguided Media? as well as other blocking obstacles do not allow two
Or short towers to communicate using microwaves,
What is Radio Transmission? Repeaters are often needed for long distance
Or communication very high frequency microwaves cannot
What is microwave use in wireless data penetrate waIls.
transmission? Parabolic dish antenna and horn antenna are used
Or for this means of transmission
What is infrared transsmission?
Ans. Unguided media transport electromagnetic
waves without using a physical conductor. [type of
communication is often referred to as wireless
communication.
There are three types of Unguided Media:
(i) Radio waves
(ii) Micro waves
(iii) Infrared.
(i) Radio waves: Electromagnetic wave ranging
in frequencies between 3 KHz and 1GHz are normally
called radio waves. Infrared: Infrared signals with frequencies ranges
Radio waves are omnidirectional when an antenna from 300 GHz to 400 1Hz can be used for short range
transmits radio waves they are propagated in all communication. Infrared signals, having high
directions. This means that sending and receiving frequencies, cannot penetrate walls. This helps to
antenna do not have to he aligned. A sending antenna prevent interference between one system and another.
can send waves that can be received by any receiving in this one room cannot be affected by the infrared waves
antenna. in another room of infrared band, almost 400 THz, has
Radio waves particularly those waves that an excellent potential for data transmission. So this will
propagate in sky mode, can travel long distances. This transfer digital data with a very high frequency. There
makes radio waves a good candidate for long-distance are no. of computer devices which are used to send the
broadcasting such as AM radio. data through infrared medium e.g. keyboard mice, PCs
Radio waves particularly those of low and medium and printers. There are some manufacturers provide a
frequencies can penetrate walls. It is an advantage special part called the IrDA port that allows a wireless
because; an AM radio can receive signals inside a keyboard to communicate with a PC.
 55

5
NETWORK AND DEVICES

Network Classifications and Topologies

so that the information is shared between different
Living in this computer age allows us to components. Wireless communication also falls in this
communicate with any person sitting is any corner of category.
the world. All this happens in just a matter of seconds. LAN OVERVIEW
Undoubtedly the credit should be given to the A local area network (LAN) supplies networking
madernization of communication technology that the capability to a group of computers in close proximity
computer has provided us. In computer terminology, to each other such as in an office building, a school, or
communication deals with transfers of data, instructions a home. A LAN is useful for sharing resources like files,
and information from one computer to another computer printers, games or other applications. A LAN in turn
through different techniques like as LAN, MAN and often connects to other LANs, and to the Internet or
WAN. This means transfer of information is achieved other WAN.
through networking, Internet, emails etc. Most local area networks are built with relatively
In this chapter, we will discuss about all the inexpensive hardware such as Ethernetcables, network
techniques and concepts of data transmission and adapters, and hubs. Wireless LAN and other more
location to another. The biggest advantage of a computer advanced LAN hardware options also exist.
network system is that a network allows its user to share Specialized operating system software may be used
programmes and data information. to configure a local area network. For example, most
flavours of Microsoft Windows provide a software
package called Internet Connection Sharing (ICS) that
NETWORK CONCEPT supports controlled access to LAN resources.
In the world of computers, networking is the The term LAN partly refers to a multiplayer gaming
practice of linking two or more computing devices event where participants bring their own computers and
together for the purpose of sharing data. Networks are build a temporary LAN.
built with a combination of computer hardware and Examples: The most common type of local area
computer software. Some explanations of networking network is an Ethernet LAN. The smallest home LAN
found in books and tutorials are highly technical, can have exactly two computers; a large LAN can
designed for students and professionals, while others accommodate many thousands of computers. Many
are geared more to home and business uses of computer LANs are divided into logical groups called subnets.
networks. An Internet Protocol (IP) “Class A” LAN can in theory
Components: The three main components of accommodate more than 16 million devices organized
networking are computer, network device and cables. into subnets.
Computers: Personal computers, servers, LAN TOPOLOGIES
mainframes etc. Think of a topology as a network’s virtual shape or
Network Device: NIC (Network Interface Card), structure. This shape does not necessarily correspond
Modem (DSL, ADSL), Hub, Switch, Bridge, Router, to the actual physical layout of the devices on the
Firewalls. network. For example, the computers on a home LAN
Cables: Cable such as copper or fibre cabling is may be arranged in a circle in a family room, but it
needed to connect the computers and networking devices would be highly unlikely to find a ring topology there.
56 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Network topologies are categorized into the topologies are found in some office buildings or school
following basic types: campuses.
bus
ring
star
tree
mesh
More complex networks can be built as hybrids of
two or more of the above basic topologies.
Bus Topology: Bus networks (not to be confused
with the system bus of a computer) use a common
backbone to connect all devices. A single cable, the
backbone functions as a shared communication medium
that devices attach or tap into with an interface
connector. A device wanting to communicate with
another device on the network sends a broadcast Ring Topology
message onto the wire that all other devices see, but
only the intended recipient actually accepts and Star Topology: Many home networks use the star
processes the message. topology. A star network features a central connection
Ethernet bus topologies are relatively easy to install point called a “hub” that may be a hub, switch or router.
and don’t require much cabling compared to the Devices typically connect to the hub with Unshielded
alternatives. 10Base-2 (“ThinNet”) and 10Base-5 Twisted Pair (UTP) Ethernet.
(“ThickNet”) both were popular Ethernet cabling Compared to the bus topology, a star network
options many years ago for bus topologies. However, generally requires more cable, but a failure in any star
bus networks work best with a limited number of
network cable will only take down one computer's
devices. If more than a few dozen computers are added
to a network bus, performance problems will likely network access and not the entire LAN. (If the hub fails,
result. In addition, if the backbone cable fails, the entire however, the entire network also fails.)
network effectively becomes unusable.

Star Topology
Bus Topology
Tree Topology: Tree topologies integrate multiple
Ring Topology: In a ring network, every device star topologies together onto a bus. In its simplest form,
has exactly two neighbors for communication purposes. only hub devices connect directly to the tree bus, and
All messages travel through a ring in the same direction each hub functions as the “root” of a tree of devices.
(either “clockwise” or “counterclockwise”). A failure This bus/star hybrid approach supports future
in any cable or device breaks the loop and can take down expandability of the network much better than a bus
the entire network. (limited in the number of devices due to the broadcast
To implement a ring network, one typically uses traffic it generates) or a star (limited by the number of
FDDI, SONET, or Token Ring technology. Ring hub connection points) alone.
 NETWORK CLASSIFICATIONS AND TOPOLOGIES / 57

methods of the cable, but two of these are commonly

used, they are
Carrier-Sense Multiple Access with Collision
Detection (CSMA/CD)
Token Passing
Carrier-Sense Multiple Access with Collision
Detection (CSMA/CD)
In this access method, each computer on the
network checks the cable for network traffic. A computer
senses that the cable(carrier-sense) is free. There are
usually many computer or nodes in a network which
Tree Topology attempts to transmit data (multiple access). This creates
a collision, as contents of the data frame will collides
Mesh Topology: Mesh topologies involve the and results in corruption of the frame. This collision
concept of routes. Unlike each of the previous will be detected by the transmitting nodes. The two
topologies, messages sent on a mesh network can take nodes involved in the transmission then wait for short
any of several possible paths from source to destination. random time interval before attempting to retransmit a
(Recall that even in a ring, although two cable paths frame once again.
exist, messages can only travel in one direction.) Some Applications
WANs, most notably the Internet, employ mesh routing. CSMA/CD was used in bus topology Ethernet
A mesh network in which every device connects to variants (is early a versions of twisted-pair
every other is called a full mesh. As shown in the figure Ethernet). In Modern Ethernet networks built
below, partial mesh networks also exist in which some with switches and/or full-duplex connections
devices connect only indirectly to others. no longer utilize CSMA/CD. The IEEE Std
802.3, which defines all Ethernet variants, for
historical reasons still bares the title “Carrier
sense multiple access with collision detection
(CSMA/CD) access method and physical layer
specifications”.
Variations of the concept are used in radio
frequency systems that rely on frequency
sharing, including Automatic Packet Reporting
System.
Token Passing: In this method, collision are
eliminated by passing a special network frame called
Token. When a computer wants to send data, it has to
Mesh Topology
wait for a free token. When it gets the token, it can
LAN ACCESS METHODS transmit the data along with the token. The receiving
If two computers put the data onto the cable at the device gets data with token, it sends the token to the
time, the data packets from one computer may collide sender as an acknowledgement. When sender receives
with another computer. If data is sent over a network the acknowledgement, the token will be released.
from one user to another or accesses from server, then Remember when token is in use by one computer the
there should be some way for the data to access the other computer cannot send data, because only one
cable without disturbing another. computer at a time can use the token. There is no
Access methods define a set of rules such as how a collision. Token passing is used by Token Ring Topology
computer puts data onto the network cable and takes and it is more expensive than CSMA/CD.
data from the cable. Access methods should be Token ring technology is a local area network
consistent in the way they handle the data. If different protocol which resides at the data link layer of the OSI
computers use different methods, the network may fail model. It is using a special three-byte frame called a
because some methods dominate the cable. Access Token that travels around the ring. Token-possession
58 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

grants the possessor permission to transmit on the A Bluetooth PAN is also called a piconet, and is
medium. Token ring frames travels completely around composed of up to 8 active devices in a master-slave
the loop. relationship. The first Bluetooth device in the piconet
Token Frame: When no station is transmitting a is the master, and all other devices are slaves that
data frame, a special token frame circles the loop which communicate with the master. A piconet typically has a
is repeated from station to station until arriving at a range of 10 meters, although ranges of up to 100 metres
station that needs to transmit data. If a station needs to
can be reached under ideal circumstances.
transmit data, it converts the token frame into a data
Local Area Network
frame for transmission. Once the sending station
receives its own data frame, it converts the frame back Local Area Network (LAN) is used to connect the
into a token. When a transmission error occurs and no devices which are geographically located very close to
token frame, or more than one, is present, a special each other, like in Home, Building itself or Campus
station referred to as the Active Monitor detects the environment. LANs are designed to allow resource
problem and removes and/or reinserts tokens as sharing between personal computer and workstation.
necessary. Shared resources can be Hardware (Printer, database),
LAN Transmission Methods: LAN transmission software (application programmes) or data. LAN
method is classified into uni-cast, multi-cast and network may be very simple depending upon the need
broadcast. In each type of transmission, a single data and type of technology used. LAN network may be just
packet is transmitted to one or more nodes. two computers and a printer, or in an organization where
“Uni-cast: A single data packet is sent from one
hundreds of computers are interconnected. In a LAN,
node to another specific node on a network. First, the
source node assigns destination node address for the you will find PCs, file server, DNS server, router, bridge,
packet. This packet is then sent to the network and finally switch, hub, gateway, multilayer switches, printers,
the packet is sent to the destination. firewall and a lot of other connecting devices. Today
“Multicast: A single data is copied and sent to LAN networks mostly used are Ethernet, Fast Ethernet
specific subset of the nodes on the network but not all (FE), Gigabit (GE) Ethernet, Token ring and FDDI. But
the networks. First, a source node addresses the each the most preferred are Ethernet and Fast Ethernet.
packet using a multicast address. This packet is then
sent to the network, which makes a copy of the packet
and sends the copy to the network to the every node
which is part of multicast address.
“Broadcast: a single data packet is send to all
nodes on the network. In this type, the source node
addresses the packet by using the broadcast address.
The packet is sent into the network which makes copies
and sends the copy to all nodes on the network.
NETWORK TYPES BASED ON SIZE
LIKE PAN, LAN, MAN, WAN
PAN: Personal Area Network (PAN) is a computer
Simple Lan
network designed for communication between computer
devices (including telephones and personal digital Wide Area Network
assistants close to one person). The devices may or may Wide Area Network (WAN) is used to connect the
not belong to the person in question. The reach of a devices which are geographically located at large
PAN is typically a few metres. PANs can be used for distances. WAN can be used to connect two or more
communication among the personal devices themselves LANs together, that may be geographically separated.
or for connecting to a higher level network and the A WAN can be as complex as the backbones that connect
Internet.Personal area networks may be wired with the internet. We have first referred to switched WAN
computer buses such as USB and FireWire. A wireless and point to point WAN.
personal area network (WPAN) can also be made Switched WAN connects the end system, which
possible with network technologies such as IrDA and usually comprises a router (internet working connecting
Bluetooth. device) that connects to another LAN or WAN.
 NETWORK CLASSIFICATIONS AND TOPOLOGIES / 59

Switched
Wan
End End
System System

End
System

The point to point WAN is normally a leased line connection which is provided by telephone or cable TV
provider that connects home computer or a small LAN to an internet service provider (ISP). This type of WAN is
used to provide an internet access.

Point to Point
Wan
Computer Modem Modem ISP

Point to Point: WAN Four basic types of connections for MANs include high-end routers, ATM
connections, or circuits, used in WAN services are: switches, and optical switches.
Circuit-switched Functional Classification
Cell-switched of Net-works
1. Peer-to-Peer: Peer-to-peer is a communications
Packet-switched
model in which each party has the same capabilities
Dedicated connections and either party can initiate a communication session.
WAN services available, including analog dialup Other models with which it might be contrasted include
are DSL (digital subscriber line) Frame Relay, ATM, the client/server model and the master/slave model. In
ISDN, X.25 and Switched Multi-megabit data services some cases, peer-to-peer communications is
(SMDS) implemented by giving each communication node both
Frame relay and X.25 are of packet switched server and client capabilities. In recent usage, peer-to-
services. peer has come to describe applications in which users
can use the Internet to exchange files with each other
Analog dialup and ISDN are circuit switched
directly or through a mediating server.
services. IBM’s Advanced Peer-to-Peer Networking (APPN)
STM and SMDS are of cell switched services. is an example of a product that supports the peer-to-
Metropolitan Area Network peer communication model.
Metropolitan Area Network (MAN) is a hybrid On the Internet, peer-to-peer (referred to as P2P)
between LAN and WAN. Like WAN, MAN usually is a type of transient Internet network that allows a group
connects two LANs in same geographical area like two of computer users with the same networking programme
buildings in a city. MAN normally covers area inside to connect with each other and directly access files from
the city. Whereas WAN usually gives low to medium one another’s hard drives. Napster and Gnutella are
speed, MAN provides high speed connection such as examples of this kind of peer-to-peer software. Major
T1 (1.54 Mbs). A good example of MAN is part of the producers of content, including record companies, have
telephone company network that can provide a high shown their concern about what they consider illegal
speed DSL line to the customer. Devices used to provide sharing of copyrighted content by suing some P2P users.
60 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Meanwhile, corporations are looking at the server at a bank. That programme may in turn forward
advantages of using P2P as a way for employees to share the request to its own database client program that sends
files without the expense involved in maintaining a a request to a database server at another bank computer
centralized server and as a way for businesses to to retrieve the account information. The balance is
exchange information with each other directly. returned to the bank database client, which in turn serves
How Does Internet P2P Work?: The user must it back to the web browser client displaying the results
first download and execute a peer-to-peer networking to the user. The client-server model has become one of
program. After launching the programme, the user enters the central ideas of network computing. Many business
the IP address of another computer belonging to the applications being written today use the client-server
network. (Typically, the Web page where the user got model. So do the Internet’s main application protocols,
the download will list several IP addresses as places to such as HTTP, SMTP, Telnet, and DNS.
begin). Once the computer finds another network The interaction between client and server is often
member on-line, it will connect to that user’s connection described using sequence diagrams. The Unified
(who has gotten their IP address from another user’s Modeling Language has support for sequence diagrams.
connection and so on). Specific types of clients include web browsers,
Users can choose how many member connections email clients, and online chat clients.
to seek at one time and determine which files they wish Specific types of servers include web servers, ftp
to share or password protect. servers, application servers, database servers, name
2. Client Server: The client-server model is a servers, mail servers, file servers, print servers, and
computing model that acts as a distributed application terminal servers. Most web services are also types of
which partitions tasks or workloads between the servers.
providers of a resource or service, called servers, and Wide Area Network: A WAN spans a large
service requesters, called clients. Often clients and geographic area, such as a state, province or country.
servers communicate over a computer network on WANs often connect multiple smaller networks, such
separate hardware, but both client and server may reside as local area networks (LANs) or metro area networks
in the same system. A server machine is a host that is (MANs).
running one or more server programs which share their The world’s most popular WAN is the Internet.
resources with clients. A client does not share any of its Some segments of the Internet, like VPN-based
resources, but requests a server's content or service extranets, are also WANs in themselves. Finally, many
function. WANs are corporate or research networks that utilize
leased lines.
WANs generally utilize different and much more
expensive networking equipment than do LANs. Key
technologies often found in WANs include SONET,
Frame Relay, and ATM.
Client Internet
WAN Topologies: A wide area network (WAN) is
a network connecting geographically distinct locations,
which may or may not belong to the same organization.
Server WAN topologies use both LAN add enterprise-wide
topologies as building blocks, but add more complexity
because of the distance they must cover, the larger
number of users they serve, and the heavy traffic they
The client-server characteristic describes the
often handle. For example, although a simple ring
relationship of cooperating programs in an application.
topology may suffice for a small office with 10 users, it
The server component provides a function or service to does not scale well and therefore cannot serve 1000
one or many clients, which initiate requests for such users. The particular WAN topology you choose will
services. depend on the number of sites you must connect, the
Functions such as email exchange, web access and distance between the sites, and any existing
database access, are built on the client-server model. infrastructure.
Users accessing banking services from their computer Peer-to-Peer: A WAN with single interconnection
use a web browser client to send a request to a web points for each location is arranged in a peer-to-peer
 NETWORK CLASSIFICATIONS AND TOPOLOGIES / 61

topology. A WAN peer-to-peer topology is similar to Star: The star WAN topology mimics the
peer-to-peer communications on a LAN in that each site arrangement of a star LAN. A single site acts as the
depends on every other site in the network to transmit central connection point for several other points. This
and receive its traffic. However, the peer-to-peer LANs arrangement provides separate routes for data between
use computers with shared access to one cable, whereas any two sites. As a result, star WANs are more reliable
the WAN peer-to-peer topology uses different locations, than the peer-to-peer or ring WANs. As a general rule,
each one connected to another one through dedicated reliability increases with the number of potential routes
circuits. data can follow. Another advantage of a star WAN is
The WAN peer-to-peer topology is often the best that when all of its dedicated circuits are functioning, a
option for organizations with only a few sites and the star WAN provides shorter data paths between any two
capability to use dedicated circuits--that is, continuously sites.
available communications channels between two access
points that are leased from a telecommunications
provider, such as an ISP.

WAN Access Methods: Internet access is a service

Ring: In a ring WAN topology, each site is that provides access to the global system of
connected to two other sites so that the entire WAN interconnected computer networks known as the
forms a ring pattern. This architecture is similar to the Internet. Consumer use first became popular through
ring LAN topology, except that a ring WAN topology dial-up connections in the 20th century. Here is a
connects locations rather than local nodes. The description of the most common Internet connection
advantages of a ring WAN over a peer-to-peer WAN methods.
are twofold: a single cable problem will not affect the Dial-Up: This method uses a modem and standard
entire network, and routers at any site can redirect data telephone line. The connection is made as necessary
to another route if one route becomes too busy. On the and the maximum speed does not exceed 56K bps.
other hand, expanding a peer-to-peer WAN because it ISDN (Integrated Services Digital Network):
requires at least one additional link. For those reasons, utilizing existing telephone lines, ISDN allows 64 Kbps
WANs that use the ring topology are only practical for on a single channel. Two channels can be combined for
connecting fewer than four or five locations. a maximum of 128 Kbps.
DSL (Digital Subscriber Line): It utilizing existing
telephone lines, DSL integrates regular phone service
and Internet access utilizing a DSL hub. This system
allows for an “always connected” situation. Speeds can
vary between 256 Kbps and 640 Kbps. A factor that my
limit your maximum speed is your physical distance
from a telephone company central office (CO).
Cable Modem: Utilizing existing cable TV coaxial
cables, this service is provided by your cable TV
provider. A cable modemdevice connects just like a TV
or cable box. The computer is connected to the cable
62 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

modem via a network card (NIC) and an Ethernet cable. Q. 4. What is the difference between DSL and
Speeds here can exceed 1.5 Mbps, but as more people cable?
in your area join the cable access network, speeds may Ans. DSL and Cable: Comparison and Contrast
diminish as everyone is sharing the same maximum 1. Speed (advantage-Cable): Cable boasts faster
bandwidth. speed than DSL internet in theory. However, cable does
T-1 - A T-1 is a very specialized telecommunications not always deliver on the promise in everyday practical
use.
circuit that does not work over normal telephone lines.
2. Popularity-(advantage-both): In the world,
It has been popular in many businesses for many years.
cable internet enjoys significantly greater popularity than
It is divided into 24 channels that can be used for many DSL, although DSL has been closing the gap recently.
different purposes, but can be combined to achieve a 3. Customer satisfaction advantage–DSL): Even
maximum speed of 1.54 Mbps. if a technology is popular, customers may be unhappy
with it whether due to cost, reliability or other factors.
4. Security (Advantage-Both): Cable and DSL
Q. 1. What is point-to-point connection? implement different network security models.
Ans. Point to point: A point-to-point connection Historically, more concerns have existed with cable
provides a dedicated link between two devices. Entire security, although cable providers have definitely taken
capacity of the link is reserved for transmission between steps to improve security over the past few years. It’s
these two devices only. likely both DSL and cable are “Secure enough” for most
people’s needs.
Q. 5. What is DSL and explain various types for
that?
Ans. Digital Subscriber Line (DSL) service is one
(Point-to-point Connection) of the most exciting technologies to come to market in
the area of information access. The concept is as elegant
Multipoint: A multipoint connection is also called as it is simple-DSL achieves broadband speeds over
as a multidrop connection. In such a connection more ordinary phone lines. That elegance and simplicity has
than two devices share a single link. attracted significant attention from a wide variety of
customers, and DSL is one of the fastest growing high-
speed access technologies for both business and personal
use on the market.
Of all of the new telecommunications technologies,
DSL is the most powerful, straightforward and flexible
solution for high-speed access and information transfer-
and the supporting network is already in place.
Multipoint Configuration Digital subscriber line is a telecommunications
Q. 2. What are the disadvantages of DSL? service that makes it possible to transform an ordinary
Ans. Disadvantages: phone line into a high-speed conduit for data, voice and
video. As long as your home or business is close enough
1. Availability: Because DSL is distance sensitive
to your service providers central office (a local office
availability is determined by the distance from the
with switching equipment which connects everyone in
provider’s central office. a certain area to the companies net-work, you’ll be able
2. Distance sensitivity limitations: Because to subscribe to DSL service. Typical connections allow
service may be limited to within 18,000 net of the DSL users to receive data at 1.5 Mbps and send data at
provides local office, or point of presence though there approximately 256 Kbps, though actual speed is
are usually several such limitations in given area. determined by the proximity to the provider’s central
Q. 3. What is RS-530? office. DSL service is always on-users don't need to dial
Ans. RS-530 is a differential communication a connection to gain access to the internet-and some
interface with some single-ended link management services even allow users to use the same line for voice
signals, typically limited to a maximum though put of and data traffic. There are wide arrays of DSL
10 Mbps. It uses DB-25 connector and differential technologies available, depending on your location and
equivalents of the V-24 signals. your bandwidth needs. In brief they are:
 NETWORK CLASSIFICATIONS AND TOPOLOGIES / 63

1. IDSL (ISDN Digital Subscriber Line) is a form Moderns are traditionally associated with PC’s in
of DSL that uses ISLJN provisioning and testing, and the form of box/PC modems,
can coexist with current analog and ISDN services. IDSL However this technology is not suited to non-PC
is usually limited to 144 Kbps upstream and equipment or “machines”, which have specialized needs.
downstream, but can sometimes provide further reach 1. Size-there is little space within many embedded
devices for a modem box
than other DSL solutions because it does not have the 2. Power consumption- some devices are battery
same distance limitations. powered and need low power modems
2. ADSL (Asymmetric Digital Subscriber Line) 3. Environment-machines can be deployed and need
employs two different transmission speeds, with the wide temperature range.
downstream speed (from the provider to the user) 4. Integration- modems need to be integrated within
usually being much higher than the upstream speed the machine and not external Modems provide remote
(from the user to the corporate host). ADSL can achieve access to machines in the field to eliminate unnecessary
downstream data rates up to 8 Mbps and upstream rates site visits and provide fast access to information in the
to I Mbps. machine. However, integrating modems.
3. VDSL (Very High Speed Digital Subscriber 1. Modems are “black art” products, surrounded
by complex compliance and regulatory issues.
Line) promises even higher speeds than ADSL, although Designing your own modem solutions requires
over much shorter distances. Standardization on speeds specialized skills. Using off-the-shelf modems enables
and technology specifications are currently in progress. designers to focus on their core application strengths
4. RADSL (Rate Adaptive Digital subscriber Line) and not be drawn into solving issues that are not related
adjusts the data transmission rate to match the quality to core competency yielding potentially unreliable
of the phone line. RADSL users get the very best products.
performance their telephone line is conditioned to M2M moderns are embedded with the machine and
provide, providing transmission rates of up to 7 Mbps transport higher level protocols between the machine
downstream and 1 Mbps upstream. and central location via the network.
5. HDSL/SDSL (High Data Rate Digital subscriber 2. Point to Point “polled” networks where pro-
prietary or industry specific protocols to communicate
Line/Symmetric Digital Subscriber Line) utilize two with central servers.
standard phone lines for 1.5 Mbps transmission speeds 3. TCP/IP enabled remote machines such as
and offer the capability to combine three phone lines Embedded PC's, where modems provide the physical
for 2.0 Mbps speeds. HDSL and SDSL are intended as network connectivity.
lower cost replacements for dedicated and fractional M2M Modems are Al-command compatible at the
T-l lines 2. serial interface to the machine and common between
Q. 6. What is a modem? Explain the types. many modems.
Ans. A modem provides the communication 4. Driven and controlled by the remote machine
interface. It transports device protocols transparently processor using industry standard Commands.
over the network through a serial interface. A modem 5. Fast time to market.
Q. 7. What is the term “Computer Networks”?
adapts the machine to communicate over various Ans. Computer network connects two or more
networks in order to gain access to the machine autonomous computers, the computers can be
including: geographically located anywhere. It provides a naturally
1. PSTN–a wire line dial-up network reductant environment of computer resources that are
2. GSM–a wireless dial-up network connected using proven communication protocols and
3. GPRS–a wireless “always on” network. reliable, high speed media.

Modem Server
Network
(GSM. GPRS) Computer

Modem Module Internet

Remote Equipment Printer

P.C.
End-to-end Protocol
64 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Q. 8. Explain different NJW Architecture?

Ans. Local area networks, metropolitan area networks, and wide area networks into which category a network
falls is determined by its size, its ownership, the distance it covers, and its physical architecture.

Categories of Network

Local Area Network (LAN): A local area network A common example of a LAN, found in many business
(LAN) is usually privately owned and links the devices environments, links a workgroup of task-related
in a single office, building, or campus (see Fig.). computers, for example, engineering workstations or
Depending on the needs of an organization and the type accounting PCs.
of technology used, a LAN can be as simple as two PCs Metropolitan-Area Network (MAN): A met-
and a printer in someone’s home office.
ropolitan-area network (MAN) is designed to extend
LANs are designed to allow resources to be shared
between personal computers or workstations. The over an entire city. It may be a single network such as a
resources to be shared can include hardware (e.g. a cable television network, or it may be a means of
printer), software (e.g., an application program), or data. connecting a number of LANs into a larger

Public City Network

Network so that the resources may be shared LAN-to-LAN as well as device-to-device. For example, a company
can use a MAN to connect the LANs in all its offices throughout a city.
Wide Area Network (WAN): A wide area network (WAN) provides long-distance transmission of data, voice,
image and video information over large geographic areas that may comprise a country, a continent, or even the
whole world.
 NETWORK CLASSIFICATIONS AND TOPOLOGIES / 65

and input on the DCE. CTS are the answering signal

coming from the DCE.
Before sending a character, the DTE asks
permission by setting its RTS output. No information
will he sent until the DCE grants permission by using
the CTS line. If the DCE cannot handle new requests,
the CTS signal will go low. A simple but useful
mechanism allowing flow control in one direction. The
assumption is that the DTE can always handle incoming
information faster than the DCE can send it in the past,
this was true. Modem speeds of 300 baud were common
Q. 9. What is Null modem? Explain the pin and 1200 baud was seen as a high speed connection.
configuration of null modem. For further control of the information flow, both
Ans. Serial communications with RS232. One of devices have the ability to signal their status to the other
the oldest and most widely spread communication side. For this purpose, the DTR data terminal ready and
methods in computer world. The way this type of DSR data set ready signals are present. The DTE uses
communication can be performed is pretty well defined the DTR signal to signal that it is read to accept
in standards i.e. with one exception. The standards show information, whereas the DCE uses the DSR signal for
the use of EYI'E/l) CE communication, the way a the same purpose. Using these signals involves not a
computer should communicate with a peripheral device small protocol of requesting and answering as with the
like a modem. For your information, DTE means data RTS/CTS handshaking. These signals are in one
terminal equipment (computers etc.) where DCE is the direction only.
abbreviation of data communication equipment The last flow control signal present in DTE/DCE
(modems). One of the main uses of serial communication communication is the CD carrier detect. It is not used
today where no modem is involved–a serial null modem directly for flow control, but mainly an indication of
configuration with DTE/DTE communication- is not so the ability of the modem device to communicate with
well defined, especially when it comes to flow control. its counter part. The signal indicates the existence of a
The terminology null modern for the situation where communication link between two modem devices.
two computers communicate directly is so often Null modem without handshaking: Flow to use
nowadays, that most people don’t realize anymore the the handshaking line in a null modem configuration?
origin of the phrase and that a null modern connection The simplest way is to don’t use them at all. In that
is an exception, not the rule. situation, only the data lines and signal ground are cross
In most situations, the original modem sign lines connected in the null modem communication cable. All
are reused to perform some sort of handshaking. other pins have no connection. An example of such a
Handshaking can increase the maximum allowed null modem cable without handshaking can be seen in
communication speed because it gives the computers the figure below.
the ability to see the flow of information. High amounts
of incoming data are allowed if the computer is capable Simple Null Modem without Handshaking
to handle it, but not if it is busy performing other tasks.
If no flow control is implemented in the null modem 1 5
connection, communication is only possible at speeds 6 9
DB 9 female

DB 9 female

2 4
7 8
at which it is sure the receiving side can handle the 8
3 3
7
4 2
amount information even under worst case conditions. 9
5 1
6
Use of RS232 When we look at the connector pin
out of the RS232 port, we see two pins which are
Connector 1 Connector 2 Function
certainly used for flow control. These two pins are RTS,
2 3 Rx Rx
request to send and CTS, clear to send. With EYI'E/ 3 2 Tx Tx
DCE communication (i.e. a computer communicating 5 5 Signal ground
with a modem device) RTS is an output on the IJTE
66

OSI and TCP/IP Models

6
Transport layer. Lower layers of the OSI model are
A set of layers and protocols is called a network represented by technologies like Ethernet. Higher layers
architecture. The specification of an architecture must of the OSI model are represented by application
contain enough information to allows a designer to write protocols like TCP and UDP.
the programme or build the hardware. The details of Need of such Models: Some of the key design
implementation and specification of the interfaces are issues that occurs in computer network are present in
hidden. A list of protocols used by a certain system, one several layers. Below we will briefly mention some of
protocol per layer, is called a protocol stack. the more important ones. Every layer needs a mechanism
The number of layers, the name of each layer, the for identifying senders and receivers. Since a network
contents of each layer, and the function of each layer normally has many computers some of which have
differ from network to network. The purpose of each multiple process, a means needed for a process on one
layer is to offer certain services to the higher layers. machine to specify with whom it wants to talk. As a
consequence of having multiple destinations, some form
INTRODUCTION OF OSI MODEL of adhering is needed in order to specify a specific
The OSI model defines internetworking in terms destination.
of a vertical stack of seven layers. The upper layers of Another set of design decisions concerns the rules
the OSI model represent software that implements for data transfer. The protocol must also determine how
network services like encryption and connection many logical channels the connection corresponds to
management. The lower layers of the OSI model and what their priorities are. Many networks provide at
implement more primitive, hardware-oriented functions least two logical channels per connection, one for normal
like routing, addressing, and flow control. data and one for urgent data.
In the OSI model, data communication starts with Error control is an important issue because physical
the top layer at the sending side, travels down the OSI communication circuits are not perfect. Many error
model stack to the bottom layer, then traveses the
detecting and error correcting codes are known but both
network connection to the bottom layer on the receiving
ends of the connection must agree on which one is being
side, and up its OSI model stack.
used. In addition, the receiver must have some way of
The OSI model was introduced in 1984. Although
telling the sender which messages have been correctly
it was designed to be an abstract model, the OSI model
received and which have not.
remains a practical framework for today’s key network
Basic Functions of each OSI Layer: The
technologies like Ethernet and protocols like IP.
International Organization for Standardization (ISO)
Examples: Internet Protocol (IP) corresponds to developed the Open Systems Interconnection (OSI)
the Network layer of the OSI model, layer three. TCP reference model to describe how information is
and UDP correspond to OSI model layer four, the transferred from one system to another system. When
 OSI AND TCP/ IP MODELS / 67

information is converted into electrical or light signals,

it is transferred along a piece of wire or radio waves. It
is very important to define a set of rules for their
transmission. Various types of rules and regulations
come under the ISO’s reference model. And it is very
important to understand that why many network
protocols, such as IP and IPX, fail to fit properly into
the scheme explained in ISO’s model.

Data Link Layer

The data link layer transforms the raw bit stream
received from the physical layer into reliable information
and is responsible for node to node delivery. It makes
the physical layer appear error free to the upper layers.
Some of the responsibilities of data layer are as
follows:
1. It deals with the physical address of both the
source and the destination.
2. When two or more devices are connected to the
same link, Data link layer protocols determine which
device will have control over the link at any given time.
This model covers all aspects of network
communication and is known as OSI model. 3. This layer also deals with error control
Physical Layer mechanism, that is, it detects and re-transmits damaged
Physical layer is required for transmitting a bit
or loss FRAME. It uses a mechanism to prevent
stream over a physical medium. It deals with mechanical
and electrical specification of the interface and duplication of frames.
transmission medium. It also defines the processor and 4. It also deals with the rate at which the sender
the functions that the physical devices and interface have
to perform for the transmission to occur, so in general should send the data to the receiver so that a fast sender
physical layer is concerned with the following: does not overwhelm a slow receiver, this mechanism is
1. Physical characteristics of interface.
called FLOW CONTROL.
2. It specifies data transmission rate.
3. It also defines the physical topologies of the 5. It divides the stream of bits received from the
network. network layer into manageable data units called
4. It deals with sender and receiver synchronization
at bit level. FRAME.
5. Representation of bits , that is, to convert digital 6. At the receiver end, again it performs the
bit into electrical signal and vice versa.
function of assembling the raw bits received from the
6. It also defines the direction of the transmission
of two devices whether it is simplex, half-duplex or full- physical layer into frames before passing them on to
duplex. the network layer.
68 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Network Layer
The network layer is responsible for source to destination delivery of packets across multiple networks. Whereas
data link layer’s delivery is hop to hop delivery. Network layer assures that each packet gets from its point of origin
to its destination.
The specific responsibilities are as follows:
1. It deals with the logical addressing of sender and receiver.
2. It deals with the routing of data between different links and networks.
3. This layer also deals with the overall congestion control mechanism of the network.
4. It divides the outgoing message into packets and assembles the incoming packets into message.
 OSI AND TCP/ IP MODELS / 69

Transport Layer 3. This layer also deals with flow control of the
The transport layer is responsible for process to data but here the flow control is performed process to
process delivery of the entire message, that is, this layer process rather than across a single link.
ensures that the whole message arrives intact and in
order over error control and flow control process to 4. It deals with service point addressing, that is,
process level. the network layer gets each packet to correct computer
Some responsibilities are as follows: and the transport layer gets the entire message to the
1. It deals with segmentation and re-assembly of correct process on the computer.
the message with proper sequence number in case the
message is very large. 5. This layer also deals with error control so that
2. It deals with connection control, that is, whether the receiving transport layer receives the message
an end to end connection should be connectionless or without any error or duplication; error correction is
connection-oriented. usually achieved by re-transmission.

Session Layer Some responsibilities are as follows:

The job of session layer is to establish, maintain 1. It deals with protocol conversion.
and synchronize the interaction between the 2. It carries out data compression to reduce the
communicating systems. bandwidth of the data to be transmitted.
Specific responsibilities of session layer are as 3. It translates data between the format that the
follows: network requires and the format that the computer
1. It allows two systems to start a communication expects.
(dialog control) with each other. The communication 4. For security and privacy purpose, it carries out
between two systems is either in HALF or FULL encryption of data at the sender’s end and decryption
DUPLEX mode. at the receiver’s end.
Application Layer
2. This layer allows addition of check points, that
This layer enables the user, whether human or
is, synchronization points into a stream of data, so that
software, to access the network. It provides user
in case of crash during the transmission of data, data
interface and support for services such as,
can be re-transmitted from the check point last inserted,
Data base access.
instead of re-transmitting from the beginning. Email.
Presentation Layer Remote file access and transfer.
The presentation layer is concerned with the syntax Some specific functions are as follows:
and semantics of information exchange between two 1. This layer allows a user to access file in a remote
communicating systems. computer.
70 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

2. This layer provides the bases of Email forwarding The Link Layer is used to move packets between
and storage. the Internet Layer interfaces of two different hosts on
3. This layer provides distributed database sources the same link. The processes of transmitting and
and access for global information about various services receiving packets on a given link can be controlled both
and objects. in the software device driver for the network card, as
4. This layer creates software emulation of a well as onfirmware or specialized chipsets. These will
terminal at remote host. This application is known as perform data link functions such as adding a packet
network virtual terminal exchange. header to prepare it for transmission, then actually
INTRODUCTION TO TCP/IP transmit the frame over a physical medium. The TCP/
The TCP/IP model (Transmission Control Protocol/ IP model includes specifications of translating the
Internet Protocol) is a descriptive framework for the network addressing methods used in the Internet
Internet Protocol Suite of computer network protocols Protocol to data link addressing, such as Media Access
created in the 1970s by DARPA, an agency of the United Control (MAC), however all other aspects below that
States Department of Defense. It evolved from level are implicitly assumed to exist in the Link Layer,
ARPANET, which was an early wide area network and but are not explicitly defined.
a predecessor of the Internet. The TCP/IP Model is This is also the layer where packets may be selected
sometimes called the Internet Model or less often the to be sent over a virtual private network or other
DoD Model. networking tunnel. In this scenario, the Link Layer data
The TCP/IP model describes a set of general design may be considered application data which traverses
guidelines and implementations of specific networking another instantiation of the IP stack for transmission or
protocols to enable computers to communicate over a reception over another IP connection. Such a connection,
network. TCP/IP provides end-to-end connectivity or virtual link, may be established with a transport
specifying how data should be formatted, addressed, protocol or even an application scope protocol that
transmitted, routed and received at the destination. serves as a tunnel in the Link Layer of the protocol stack.
Protocols exist for a variety of different types of
Thus, the TCP/IP model does not dictate a strict
communication services between computers.
hierarchical encapsulation sequence.
Layers in the TCP/IP Model: The layers near the
Internet Layer: The Internet Layer solves the
top are logically closer to the user application, while
problem of sending packets across one or more
those near the bottom are logically closer to the physical
networks. Internetworking requires sending data from
transmission of the data. Viewing layers as providing
the source network to the destination network. This
or consuming a service is a method of abstraction to
process is called routing.
isolate upper layer protocols from the nitty-gritty detail
In the Internet Protocol Suite, the Internet Protocol
of transmitting bits over, for example, Ethernet and
performs two basic functions:
collision detection, while the lower layers avoid having
to know the details of each and every application and Host addressing and identification: This is
its protocol. accomplished with a hierarchical addressing
The following is a description of each layer in the system.
TCP/IP networking model starting from the lowest level: Packet routing: This is the basic task of
Physical Layer: The Physical Layer is pure getting packets of data (datagrams) from
hardware in any network infrastructure. This includes source to destination by sending them to the
the cable, satellite or any other connection medium, and next network node (router) closer to the final
the network interface card, which transmits electrical destination.
signals, and so on. IP can carry data for a number of different upper
Link Layer: The Link Layer (or Network Access layer protocols. These protocols are each identified by
Layer) is the networking scope of the local network a unique protocol number: for example, Internet Control
connection to which a host is attached. This regime is Message Protocol (ICMP) and Internet Group
called the link in Internet literature. This is the lowest Management Protocol (IGMP) are protocols 1 and 2,
component layer of the Internet protocols, as TCP/IP is respectively.
designed to be hardware independent. As a result TCP/ Some of the protocols carried by IP, such as ICMP
IP is able to be implemented on top of virtually any (used to transmit diagnostic information about IP
hardware networking technology. transmission) and IGMP (used to manage IP Multicast
 OSI AND TCP/ IP MODELS / 71

data) are layered on top of IP but perform “unreliable” protocol. Reliability is addressed through
internetworking functions. This illustrates the error detection using a weak checksum algorithm. UDP
differences in the architecture of the TCP/IP stack of is typically used for applications such as streaming
the Internet and the OSI model. media (audio, video,Voice over IP etc.) where on-time
Transport Layer: The Transport Layer ’s arrival is more important than reliability, or for simple
responsibilities include end-to-end message transfer query/response applications like DNS lookups, where
capabilities independent of the underlying network, the overhead of setting up a reliable connection is
along with error control, segmentation, flow control, disproportionately large. Real-time Transport Protocol
congestion control, and application addressing (port (RTP) is a datagram protocol that is designed for real-
numbers). End-to-end message transmission or time data such as streaming audio and video.
connecting applications at the transport layer can be TCP and UDP are used to carry an assortment of
categorized as either connection-oriented, implemented higher-level applications. The appropriate transport
in Transmission Control Protocol (TCP), or protocol is chosen based on the higher-layer protocol
connectionless, implemented in User Datagram Protocol application. For example, the File Transfer Protocol
(UDP). expects a reliable connection, but the Network File
The Transport Layer can be thought of as a transport System (NFS) assumes that the subordinate Remote
mechanism, e.g. a vehicle with the responsibility to make Procedure Call protocol, not transport, will guarantee
sure that its contents (passengers/goods) reach their reliable transfer. Other applications, such as VoIP, can
destination safely and soundly, unless another protocol tolerate some loss of packets, but not the reordering or
layer is responsible for safe delivery. delay that could be caused by retransmission.
The Transport Layer provides this service of The applications at any given network address are
connecting applications through the use of service ports. distinguished by their TCP or UDP port. By convention
Since IP provides only a best effort delivery, the certain well known ports are associated with specific
Transport Layer is the first layer of the TCP/IP stack to applications.
offer reliability. IP can run over a reliable data link Application Layer: The Application Layer refers
protocol such as theHigh-Level Data Link Control to the higher-level protocols used by most applications
(HDLC). Protocols above transport, such as RPC, also for network communication. Examples of application
can provide reliability. layer protocols include the File Transfer Protocol (FTP)
For example, the Transmission Control Protocol and the Simple Mail Transfer Protocol (SMTP).Data
(TCP) is a connection-oriented protocol that addresses coded according to application layer protocols are then
numerous reliability issues to provide a reliable byte encapsulated into one or (occasionally) more transport
stream: layer protocols (such as the Transmission Control
data arrives in-order Protocol(TCP) or User Datagram Protocol (UDP)),
data has minimal error (i.e. correctness) which in turn use lower layer protocols to effect actual
duplicate data is discarded data transfer.
lost/discarded packets are resent Since the IP stack defines no layers between the
includes traffic congestion control application and transport layers, the application layer
The newer Stream Control Transmission Protocol must include any protocols that act like the OSI’s
(SCTP) is also a reliable, connection-oriented transport presentation and session layer protocols. This is usually
mechanism. It is Message-stream-oriented–not byte- done through libraries.
stream-oriented like TCP–and provides multiple streams Application Layer protocols generally treat the
multiplexed over a single connection. It also provides transport layer (and lower) protocols as “black boxes”
multi-homing support, in which a connection end can which provide a stable network connection across which
be represented by multiple IP addresses (representing to communicate, although the applications are usually
multiple physical interfaces), such that if one fails, the aware of key qualities of the transport layer connection
connection is not interrupted. It was developed initially such as the end point IP addresses and port numbers.
for telephony applications (to transport SS7 over IP), As noted above, layers are not necessarily clearly
but can also be used for other applications. defined in the Internet protocol suite. Application layer
User Datagram Protocol is a connectionless protocols are most often associated with client-server
datagram protocol. Like IP, it is a best effort, applications, and the commoner servers have specific
72 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

ports assigned to them by the IANA: HTTP has port IP contributed to OSI, and vice-versa. Several important
80; Telnet has port 23; etc. Clients, on the other hand, differences do exist, though, which arise from the basic
tend to use ephemeral ports, i.e. port numbers assigned requirements of TCP/IP which are:
at random from a range set aside for the purpose. A common set of applications
Transport and lower level layers are largely Dynamic routing
unconcerned with the specifics of application layer Connectionless protocols at the networking
protocols. Routers and switches do not typically “look level
inside” the encapsulated traffic to see what kind of Universal connectivity
application protocol it represents, rather they just Packet-switching
provide a conduit for it. However, some firewall and The main differences between the OSI architecture
bandwidth throttling applications do try to determine and that of TCP/IP relate to the layers above the transport
what’s inside, as with the Resource Reservation Protocol layer (layer 4) and those at the network layer (layer 3).
(RSVP). It’s also sometimes necessary for Network OSI has both, the session layer and the presentation
Address Translation (NAT) facilities to take account of layer, whereas TCP/IP combines both into an application
the needs of particular application layer protocols. (NAT layer. The requirement for a connectionless protocol also
allows hosts on private networks to communicate with required TCP/IP to combine OSI’s physical layer and
the outside world via a single visible IP address using data link layer into a network level.
port forwarding, and is an almost ubiquitous feature of Physical Layer: The physical layer may be either
modern domestic broadband routers). ethernet, SDH-DCC, or some timeslot of a PDH signal.
Either OSI protocols and TCP/IP protocols build on the
same physical layer standards, thus there is no difference
between OSI and TCP/IP in this aspect.
Data Link Layer: The purpose of the data link
layer is to provide error free data transmission even on
noisy links. This is achieved by framing of data and
retransmission of every frame until it is acknowledged
from the far end, using flow control mechanisms. Error
detection is done by means of error detection codes.
The data link layer in the OSI world makes use of
the Q.921 LapD protocol which must support an
information field length of at least 512 octets according
to G.784. LapD is based on HDLC framing.
In the internet world there is no real data link layer
protocol, but the subnet protocol which has quite many
similarities. The subnet protocol consists of the IMP-
IMP protocol which aims to provide a reliable
connection between neighboured IMPs.
For ethernet based networks e.g. LANs (Local Area
Network), the data link protocol LLC (Logical Link
Control) is equally used in OSI and TCP/IP networks.
Network Layer: The network layer provides
routing capabilities between source and destination
COMPARISONS WITH TCP/IP LAYERS system.
AT THE BEGINNING’S LEVEL OSI uses the CLNS (Connection Less Network
This chapter gives a brief comparison between OSI Service) protocols ES-IS for communication of an end
and TCP/IP protocols with a special focus on the system to an intermediate system and IS-IS for
similarities and on how the protocols from both worlds communication between intermediate systems.
map to each other. The adoption of TCP/IP does not TCP divides messages in datagrams of up to 64k
conflict with the OSI standards because the two protocol length. Each datagram consists of a header and a text
stacks were developed concurrently. In some ways, TCP/ part. Besides some other information, the header
 OSI AND TCP/ IP MODELS / 73

contains the source and the destination address of the basic layers which can be related to seven layers of ISO-
datagram. IP routes these datagrams through the network OSI reference model.
using e.g. the protocol OSPF (Open Shortest Path First) According to TCP/IP reference model total
or RIP (Route Information Protocol) for path calculation functionality of communication process is divided into
purposes. The service provided by IP is not reliable. four different layers:
Datagrams may be received in the wrong order or they 1. Internet layer
may even get lost in the network. 2. Transport layer
Transport Layer: The transport layer provides a 3. Application layer
reliable end-to-end connection between source and 4. Host to network layer.
destination system on top of the network layer. It builds 1. The Internet Layer: This layer holds the whole
an integral part of the whole OSI layering principle and architecture together. Its job is to permit hosts to inject
of the internet protocol. packets into any network and have they travel
The OSI transport layer protocol (TP4) and the independently to the destination. They may arrive in a
internet tranport protocol (TCP) have many similarities different order then they were sent, in which case it is
but also some remarkable differences. Both protocols the job of higher layers to rearrange them, if in-order
are built to provide a reliable connection-oriented end- delivery is desired.
The internet layer defines an official packet format
to-end transport service on top of an unreliable network
and protocol called internet protocol (IP). The job of
service. The network service may loose packets, store
internet layer is to deliver IP packets where they are
them, deliver them in the wrong order or even duplicate
supposed to go. Packet routing is clearly the major issue
packets. Both protocols have to be able to deal with the
here, as is avoiding congestion.
most severe problems e.g. a subnetwork stores valid 2. Transport Layer: The layer above the internet
packets and sends them at a later date. TP4 and TCP layer is called transport layer. It is designed to allow
have a connect, transfer and a disconnect phase. The peer entities on the source and destination hosts to carry
principles of doing this are also quite similar. on a conversation, just as in the OSI transport layer.
One difference between TP4 and TCP to be Two end to end transport protocols have been defined
mentioned is that TP4 uses nine different TPDU here. The first one TCP, is a reliable connection-oriented
(Transport Protocol Data Unit) types whereas TCP protocol that allows a byte stream originating one
knows only one. This makes TCP simpler but every TCP machine to be declined without error on any other
header has to have all possible fields and therefore the machine in the internet.
TCP header is at least 20 bytes long whereas the TP4 The second protocol in this layer, UDP (User
header takes at least 5 bytes. Datagram Protocol), is an unreliable, connectionless
Another difference is the way both protocols react protocol for applications that do not want TCP’s
in case of a call collision. TP4 opens two bidirectional sequencing or flow control and wish to provide their
connections between the TSAPs whereas TCP opens own, It is also used for one-short, client-server-type
just one connection. request reply queries and applications in which prompt
TP4 uses a different flow control mechanism for delivery is more important than accurate delivery, such
its messages, it also provides means for quality of service as transmitting speech or video.
measurement. 3. The Application Layer: The TCP/IP model
does not have session or presentation layers. No need
for them was perceived, so they were not included.
Q. 1. Explain TCP/IP Protocol. Experience with an 051 model has proven this view
Ans. This is most widely used protocol for networks correct, they are of little be use to most applications.
throughout the world. This technology is adopted as the On top of transport layer is the application layer. It
global standard of networking. This is combination of contains all the higher-level protocols. The early ones
two basic protocols the transmission control and internet. included virtual terminal (TELNET), file transfer (FTP),
This technology is used by the internet, internet and and electronic mail (SMTP). The virtual terminal
extranets. Many operating systems and networking protocol allows a user on one machine to log on to a
software are coming up with compatibility to this distant machine and work their. The file transfer protocol
architecture. Currently most of the organizations are provides a way to move data efficiently from one
adopting this technology. This protocol consists of five machine to another.
74 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

4. The Host to Network Layer: Below internet Q. 3. What is the difference between physical
layer is a great void. The TCP/IP model does not really address and logical address?
by much about what happens here, except to point out Ans. A Physical address is a 48-bit flat address
that the host has to connect to the network using some burned into the ROM of the NIC card which is a Layer1
protocol so it can send IP packets to it. device of the OSI model. This iis divided into 24-bit
This protocol is not defined and varies from host vendor code and 24-bit serial address. This is unique
to host and network to network. for each system and cannot be changed.
Q. 2. Write a short note on Ethernet. A Logical address is a 32- bit address assigned to
Ans. Ethernet is the most widely used LAN each system in a network. This works in Layer-3 of OSI
Protocol. The original Ethernet was created in 1976. Model. This would be generally the IP address.
Ethernet is designed to operate at 10 Mbps. Access to Q. 4. What are transmission impairments?
the network by a device is through a contention method Explain all of them.
(CSMA/CD). The Ethernet frame contains seven fields. Ans. When transmission takes place between
1. Preamble media, which is not perfect. The imperfections cause
2. Start frame delimiter (SFD) impairment in the signal. This means that the signal at
the beginning and end of the medium art not the same.
3. Destination address (DA)
There are chances for the data that we have sent and we
4. Source address
have received. There are three types of impairment
5. Length/type usually occur
6. Data 1. Attenuation
7. CRC 2. Distortion
Each station on an Ethernet network has its own 3. Noise
network interface card (NIC). The NIC fits inside the
station and provides the station with a 6 byte physical
address. The Ethernet address is 6 bytes (48 bits) that is
normally written in hexadecimal notation. In Ethernet a
source address is always a unicast address and the
destination address can be unicast, multicast or
broadcast. There are the different categories of
traditional 1. Attenuation: Attenuation means loss of energy
10 Base-T is Twisted-pair Ethernet. when a signal travels through the mediums, it losses
Base-3 FL is the fiber link Ethernet. some of its energy so that it can overcome the resistance
Thick Ethernet uses a bus topology with an external of the medium. That is why a wire carrying electrical
transceiver connected via a tap to a thick coaxial cable. signals gets warm to compensate for this loss, amplifiers
This Ethernet uses a bus topology with an internal are used to amplify the signal.
transceiver or a point-to-point connection via an external
transceiver.10 base-T uses a physical star topology. The
stations are connected to a hub with an internal
transceiver or an external transceiver.
10 Base FL uses a star topology to connect stations
to a hub. The standard is normally implemented using
an external transceiver called fibre optic MAV.
LAN is divided into Bridges. A bridge can raise
the bandwidth and separate the collision domains on an
Ethernet LAN. A switch allows each station on an
Ethernet. Distortion: Distortion means that the signal
LAN to have the entire capacity of the network to changes its forms or shape. Distortion occurs in a
itself. composite signal, made of different frequencies. Each
 OSI AND TCP/ IP MODELS / 75

signal component has its own propagation speed through medium and therefore its own delay is arriving at the final
destination. Fig. shows the effect of distortion on composite signal.

Noise: Noise also causes problem in the Q. 5. Explain the three layer model briefly.
transmission process. Several types of noise such as Ans.We can organize the communication task into
thermal noise, induced noise crosstalk and impulse noise three independent layers:
may corrupt the signal. 1. Network Access Layer: It is concerned with
exchange of data between a computer and the network
to which it is attached. The sending computer must
provide the network with the address of destination
computer, so that the network may route the data to the
appropriate destination. The specific software used at
the layer depends on the type of network to be used.
Different standards have been developed for circuit
switching, packet switching, LAN’s and others.
Thermal noise is the random motion of electrons 2. Transport Layer: Regardless of nature of
in a wire which creates an extra signal not originally applications that are exchanging data, there is usually a
sent by the transmitter. requirement that data be exchanged reliably. That is we
Induced noise causes from sources such as motors would like to be assured that all of the data arrive at the
and appliances. These devices out as a sending antenna destination application and that the data arrived in the
and the transmission medium acts as the receiving same order in which they were sent. The mechanisms
antenna. for providing reliability are independent of nature of
Crosstalk is the effect of one wire on the other. One applications. Thus, it makes sense to correct those
wire act as sending antenna and other as the receiving mechanisms in a common layer shared by all
antenna. applications this is referred to as transport layer.
Impulse noise is a spike for a very short period of 3. Application Layer: Application layer contains
time that comes from power lines, lightning and so on. the logic needed to support the various user applications.
For each different type of application such as file
transfer, a separate module is needed that is peculiar to
that application.
Q. 6. Define service access points?
Ans. Each application on a computer must have a
unique network address. This allows the network to
deliver data to the proper computer. Each application
76 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

on a computer must have an address that is unique within of the organization. The network which connects the
that computer. This allows the transport layer to support various locations and gives connectivity within the
multiple applications at each computer. These latter organization is called intranet. These networks are
addresses are known as service access points (SAPs) limited to the organization for which they are set up.
Q. 7. What is PDU (Protocol Data Unit)? These intranets can be connected to other intranets and
Ans. In order to transfer the SDU, the layer n entity extranets forming internet.
has to divide it into many smaller pieces. Each piece is Intranets are designed to be open with security,
given a header and sent as a separate PDU such as a whose web browsing to the end users. There internal
packet. websites may be designed by company, its business,
The PDU headers are used by the peer entities to units, departments, workgroups etc. For example, a
carry out their peer protocol. Some PDUs contain data marketing department may design an intranet website
while other PDUs contain the control information. The so that anyone can know about their products and,
PDU headers will identify or differentiate between services an. have detail information about them.
different types of PIJUs. They also provide sequence Q. 11. What is Extranet?
numbers and counts. Ans. There are those networks, which link some
Q. 8. What are the merits and demerits of OSI of the internets of the company with those of its business
Reference model? partners, customers, suppliers, consultants, etc. who cart
Ans. Merits: access selected intranet websites and company’s
1. The protocols in OSI model are better hidden. databases.
So they can be easily replaced by new protocols as the Most of the organizations establish private extranets
technology changes. among themselves, so that they can be real business
2. OSI model is truly a general model. partners, who share information, resources etc. Instead
3. It distinguishes very clearly between the services, of private extranets, organizations can also use the
interfaces and protocols.
internet, to establish network connections between them.
4. This, model supports connection oriented as well
Some organizations use virtual private networks (VPNs)
as connectionless services.
to establish secure Intranets and Extranets.
Demerits:
Q. 12. Name any five standard creation
1. Sessions and presentation layers are not of much
committees.
use.
Ans. 1. International Organization for Standardi-
2. This model was devised before the protocols
were invented. So in real life there is a problem of fitting zation (ISO)
protocol into model. 2. International Telecommunication Union-
Q. 9. Define Internet? Telecommunication Standards
Ans. As the business operations are becoming, 3. American National Standards Institute (ANSI)
global, where one has to have connectivity with global 4. Institute of Electrical and Electronics Engineers
customers, other organization etc. This connectivity is (IIFl)
done by connecting many LANs using wide area 5. Electronic Industries Association (ElA).
networks. These inter connected networks are generally Q. 13. Describe connection-oriented and
known as inter network or popularly known as Internet. connectionless services.
We can define internet as a network of networks, which Ans. Layers can offer two different types of service
connects different organizations, to share information, i.e. connection-oriented and connectionless services.
resource etc. These networks enable end users to Connection-oriented Service: It is modelled after
communicate with other users on other LANs, so that the telephone system. To talk to someone, you pick up
one can access the computing resources, organizations. the phone, dial the number, talk and then hang up.
These networks for their connectivity rely on inter Similarly, to use a connection oriented network system,
networks processors like switches, routers hub gateways the service user first establishes a connection, uses the
etc. connection, and then releases the connection. The
Q. 10. Write a short note on Intranet. essential aspect of a connection is that it act like a tube,
Ans. The basic goal of many organizations is to the sender pushes objects in at one end, and receiver
create connectivity between each and every branches takes them out in same order at the other end.
 OSI AND TCP/ IP MODELS / 77

Connectionless Service: It is modelled after the have to be overcome, by this layer to allow heterogenous
postal system. Each message carries the lull destination network to be interconnected.
address and each one is routed through the system 4. Transport Layer: This layer helps organization
independent of all others. Normally, when two messages and transfer of data between the nodes. The basic
are sent to the same destination, the first one sent will function of this layer to accept the data from the session
be the first one to arrive. layer divide it into smaller’ units, if needed, then send
Q. 14. Explain the OSI Reference model. them to the network layer. This layer ensures that all
Ans. This standard model for network protocol was the data pieces arrive correctly, at the receiving end.
developed by international standards organization (ISO) This layer also determines the type of service to be
which was the major step in standardization on network provided to the session layer.
communications. This model is known as ISO-OSI 5. Session Layer: This layer is concerned with
(International standards organization – open system establishment of sessions between two computers to
interconnection) reference model, as it deals with establish connection. A session transmits the data like,
connecting open system i.e. the systems which are open transport layer, but it provides useful enhanced service.
for communication with other systems. This model This layer allows establishing connection to remote
serves as a standard model for network architecture. systems. This layer determines which user has the turn
OSI model is divided into seven layers as follows: for data transmission. It does it using token management
1. Physical Layer: This layer is concerned with technique, where tokens are given to data, the data which
the transmission of raw bits, over the communication possess the token can do a data transmission.
channel. This layer provides physical transmission of 6. Presentation Layer: This layer provides data
data of the communication media in the network. While transmission codes and formats. This layer is concerned
designing this layer, it has to be made sure that when with syntax and semantics of the information which is
one computer sends a 1 bit, other competitor should being transmitted. This layer handles the abstract data
receive it’ as a I bit, not only as a 0 bit. Here, the structures and conversion to the standard representation.
concerned issues largely deal with mechanical, electrical The data encryption compression etc. is used for data
and procedural interfaces. This layer is in the domain transmission, whose representation is taken care of by
of electrical engineer. this presentation layer.
2. Data Link Layer: This layer is concerned with 7. Application Layer: This layer provides
providing on error free Organization and transmission communication services for user applications. This layer
of data into data frames transmit these frames contains a large number of protocols which are
sequentially and process the acknowledgement frames
commonly required transmission of data. All the
which are sent back by the receiver. As the physical
software required for interacting with incompatible
layer is concerned only with accepting and, transmission
hardware and software are available in this layer. This
of bits without knowing structure of them, it is up to
layer is also used for file transfer. This layer provides
data link layer to create and recognize the frame
all general purpose and special purpose facilities.
boundaries.
Q. 15. Write down the basic principles of OSI
This task is done by attaching special bit patterns
to the beginning and end of frames. model.
3. Network Layer: This layer provides appropriate Ans. 1. Whenever, there is a need for different
routing by establishing connections among the various levels of data abstraction, a different layer has to be
subnets. This layer design is concerned with created.
determination of routing of data packets across the 2. Every layer has well defined functions.
source to the destination. When many packets are 3. The functions of every layer should be based on
present in the subnet, it creates congestion which has to international standard protocols.
be controlled by this layer. 4. The layers should have boundaries to minimize
When data has t be routed across different networks, the flow of information across the interfaces.
the addressing of the data packet destinations, which 5. The total number of layers should be large enough
poses great problems. The difference in protocols of so that functions which are distinct should not be
networks may also pose problems. All these problems performed by one layer.
78

Physical and Data Link Layer

7
than the number of bits that would be required to do an
Networks must be able to transfer data from one error correcting code. This approach will lead to adding
device to another with complete accuracy. A system that checksums or cyclic redundancy checks to the string.
cannot guarantee that the data received by one device The Hamming distance between two bit strings is
are identical to the data transmitted by another device the number of bits you have to change to convert one to
is essentially useless. Yet, anytime data are transmitted the other: this is the same as the number of edges you
from source to destination they can become corrupted have to traverse in a binary hypercube to get from one
in passage. In fact, it is more likely that some part of a of the vertices to the other. The basic idea of an error
message will be altered in transit than that the entire correcting code is to use extra bits to increase the
contents will arrive intact. Many factors including line dimensionality of the hypercube, and make sure the
noise can alter or wipe out one or more bits of a given Hamming distance between any two valid points is
data unit. Reliable systems must have a mechanism for greater than one.
detecting and correcting such errors. If the Hamming distance between valid strings
is only one, a single-bit error results in another
valid string. This means we can’t detect an error.
ERROR DETECTION AND CORRECTION If it’s two, then changing one bit results in an
The basic problem we have to resolve is that invalid string, and can be detected as an error.
memory and communications technology isn’t totally Unfortunately, changing just one more bit can
reliable; we have to expect and be ready to deal with result in another valid string, which means we
errors in the hardware. This document will describe two can’t know which bit was wrong: so we can
very different technologies for detecting and maybe detect an error but not correct it.
correcting, errors that may occur in data storage and If the Hamming distance between valid strings
transmission. is three, then changing one bit leaves us only
The first approach to be described is more one bit away from the original error, but two
appropriate for environments like memory: a relatively bits away from any other valid string. This
small amount of data is fetched in parallel. This means if we have a one-bit error, we can figure
approach, called “error detecting and correcting codes,” out which bit is the error; but if we have a two-
is based on defining a distance between two bit strings bit error, it looks like one bit from the other
in terms of the number of bits that have to change to get direction. So we can have single bit correction,
from the first string to the second. Extra bits are added but that’s all.
to each string, which are set so that some minimum Finally, if the Hamming distance is four, then
number of bits must change to get from one valid string we can correct a single-bit error and detect a
to another. If the received string isn’t valid, it is assumed double-bit error. This is frequently referred to
that the correct string is the one “closest” to the received as a SECDED (Single Error Correct, Double
string. Error Detect) scheme.
The second approach is more appropriate to Parity: The simplest case is by adding a parity bit.
environments in which relatively large amounts of data Suppose we have a three-bit word (so the bit strings
are to be transferred, but they are transferred serially. define points in a cube). If we add a fourth bit, we can
In this approach a “signature” is appended to the data decree that any time we want to switch a bit in the
string; the number of bits in the signature is much less original three-bit string, we also have to switch the parity
 PHYSICAL AND DATA LINK LAYER / 79

bit. If we start with 000 in the left cube, so the full string all contain 20; those data bits are used in calculating
is 0000, changing any one of the original three bits check bit C1. We simply set C1 as having the parity of
requires us to change to the other cube: 1001, 1010, its data bits.
and 1100. Now if we change a second bit, we have to Looking at all the check bits, we get:
move back to the left cube: 0011, 0101, 0110. And if C1 = M1 ^ M2 ^ M4 ^ M5 ^ M7
we change the third bit, we move back to the right cube: C2 = M1 ^ M3 ^ M4 ^M6 ^ M7
0111. C4 = M2 ^ M3 ^ M4 ^ ^ M8
So, there is a Hamming distance of two between C8 = ^ M5 ^ M6 ^ M7 ^ M8
any two valid strings. If we get a one-bit error, we know Now, if we get an error, the parity will be wrong
it is an error because it’s on one of the invalid vertices. for all of the sets based on that bit. The check bits that
This can be computed by counting the number of turn up wrong will be the bit number of the error!
1’s, and making sure it’s always even (so this is called We can combine ECC with parity. The way we do
even parity). We could have selected exactly the this, is we take the parity over all the bits in the word
opposite set of vertices as the valid ones, which would (including the check bits). In our bit numbering scheme,
have given us odd parity. We picked even parity because we consider Parity as bit 0000.
we’ll be using it in the next step. So, when we look at the parity and check bits, we
Error Correction: The weakness of the parity get the following results:
scheme is that we can tell we had an error, but we can’t If the parity is correct and the check bits are
know which bit is wrong. If we use enough extra bits, correct, our data is correct.
we can tell not only that a bit is wrong, but which one it If the parity is incorrect, the check bits indicate
is. Since we need to have enough check bits to spot which bit is wrong. If the check bits indicate
both an error in the data and in the check bits themselves that the error is in bit 0000, it’s the parity bit
(after all, they aren’t going to be perfect either), we need itself that is incorrect.
(log n) + 1 bits (Hamming derives this result much, much If the parity is correct but the check bits indicate
more carefully in his paper). The basic idea in what an error, there is a two-bit error. This can’t be
follows is that we’ll divide the data bits into log n subsets corrected.
where each subset contains roughly half of all the bits, CYCLIC REDUNDANCY CHECKS
and compute the even parity of each subset. If we have As an example, let’s set a 5-bit polynomial of
an error, we’ll be able to tell which bit has the error 11001, and compute the CRC of a 16 bit message:
because it will be uniquely determined by the set of
subsets that turn up with bad parity.
Bit Position Check Data
Position Number Bit Bit
12 1100 M8
11 1011 M7
10 1010 M6
9 1001 M5
8 1000 C8
7 0111 M4
6 0110 M3
5 0101 M2
4 0100 C4
3 0011 M1
2 0010 C2
1 0001 C1
Here’s how we find the subsets: The data bit
positions which contain a 1 in the bit corresponding to When we did the division, we didn’t bother to keep
a check bit number are used in calculating that check track of the quotient; we don’t care about the quotient.
bit. So, looking at the table, data bits M1, M2, M4, M5, Our only goal here is to get the remainder (0101), which
and M7 are in rows 3, 5, 7, 9, and 11; those row numbers is the FCS.
80 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

The key insight is that we can perform a subtraction polynomial written in polynomial form; the one we’ve
any time there is a 1 in the bit that lines up with the most been using would be written as x4 + x3 + x0.
significant bit of the polynomial, and we can perform So, just a little bit more. First, there is quite a bit of
that subtraction by performing an exclusive–or of the theory behind choosing a “good” CRC polynomial; the
bits corresponding to 1’s in all the other places of the choice of polynomial can be tuned to make sure that
polynomial. This lets us implement the CRC calculation any burst of some given length can be caught.
by using a shift register similar to the one for vertical Properties of Cyclic Redundancy Checks: The
parity. paper lists a few properties of CRCs, which deserve
You can see how it’s done by comparing the division mention:
we performed above to the circuit in the next figure. If the rightmost place of the generating
The figure shows a shift register; the string to be checked polynomial were 0, the generating polynomial
is inserted from the right. Whenever a “1” exits the left would be divisible by X. That being the case,
side of the shift register, it means there is a 1 in the most any polynomial divisible by P would also be
significant bit of the part of the dividend we’re working divisible by X, and so the last bit of the check
with; since we’re working in modulo-2 arithmetic, this bits would always be 0. That would be useless,
means we can do a subtraction. What this works out to so we always have a 1 in the least significant
is: bit of the generating polynomial.
1. The most significant bit will be xored away, so That’s a roundabout way of saying that if you’re
it falls off to the left. going to have an n–k bit polynomial, the two
2. For every other bit with a “1” in the divisor, outlying bits should be 1’s, otherwise you’ve
perform an exclusive-or with the corresponding effectively got a shorter polynomial than that.
bit in the number being checked. Any error checking code that can always detect
3. For bits with a “0” in the divisor, do nothing. a two-bit error can always correct any one-bit
The figure below attempts to show this for the error. In the most ridiculous case, we can just
example CRC polynomial. Each of the square boxes is check by flipping every bit of the received
a position in the shift register, where a value can be message; whenever we flip the wrong bit we
stored. Every round box is a position where we may or get a two-bit error, when we flip the right one
may not perform an exclusive–or, depending on the we get a 0-bit error. Of course, Hamming’s
polynomial we’re using. You can see the value of the scheme is a lot more clever than this!
CRC polynomial written above the round boxes. Any cyclic code whose generating polynomial
is of length n–k will always detect any burst
error of length less than n–k.
There are a few “classic” CRC polynomials of given
I keep calling this a polynomial, and writing it as a lengths which are so sell established that they’ve been
binary number. Frequently, you’ll find a CRC given names.

Name Definition
As Polynomial As Number
CRC12 X12 + X11 + X3 + X + 1 1100000001011
CRC16 X16 + X15 + X2 + 1 11000000000000101
CRC-CCITT X16 + X12 + X5 + 1 10001000000100001
CRC32 X32 +X26 + X23 + X22 +X16 + X12 + 100000100110000010001110110110111
X11 +X10 + X8 + X7 +X5 + X4 + X2 + X + 1
CRC
The most powerful of the redundancy checking techniques is the cyclic redundancy check (CRC). CRC is
based on binary division. In CRC instead of adding bits together to achieve a desired parity a sequence of redundant
bits, called the CRC or the CRC remainder, is appended to the end of a data unit so that the resulting data unit
becomes exactly divisible by a second predetermined binary number. At its destination the incoming data unit is
 PHYSICAL AND DATA LINK LAYER / 81

divided by the same number if at this step there is no clock reference. SONET frames are then
remainder the data unit is assumed to be intact and is positioned within the clocked stream.
therefore, accepted. A remainder indicates that the data RETRANSMISSION STRATEGIES
unit has been damaged in transit and therefore, must be Flow control is a set of procedures that tell the
rejected. sender how much data it can transmit before it must
FRAMING wait for an acknowledgement from the receiver. Two
A point-to-point connection between two computers points can be made in this context:
or devices consists of a wire in which data is transmitted The flow of data cannot be allowed to
as a stream of bits. However, these bits must be framed overwhelm the receiver. Any receiving device
into discernible blocks of information. Framing is a has a limited speed to process incoming data,
function of the data link layer. It provides a way for a and a limited memory to store it. It must be
sender to transmit a set of bits that are meaningful to able to tell that the transmitting device to send
the receiver. Ethernet, token ring, frame relay, and other a fewer frames or stop temporarily. Incoming
data link layer technologies have their own frame data has to be checked and processed before
structures. Frames have headers that contain information use. The rate of such processing is often slower
such as error-checking codes. than the rate of transmission. Each receiving
There are three different types of framing, each of device thus has a block of memory, called a
buffer, to store incoming data until they are
which provides a way for the sender to tell the receiver
processed. When the buffer begins to fillup, the
where the block of data begins and ends:
receiver has to tell the sender to stop
Byte-oriented Framing: Computer data is
transmission till it is once again able to receive.
normally stored as alphanumeric characters that
As frames come in, they are acknowledged,
are encoded with a combination of 8 bits either frame by frame or several frames at a
(1 byte). This type of framing differentiates one time. If a frame arrives damaged, the receiver
byte from another. It is an older style of framing sends an error message (a NAK frame).
that was used in the terminal/mainframe Flow control, thus, refers to a set of procedures
environment. Examples of byte-oriented used to restrict the amount of data the sender can send
framing include IBM’s BISYNC protocol. before waiting for acknowledgement.
Bit-oriented Framing: This type of framing There are two basic methods to control the flow of
allows the sender to transmit a long string of data on communications links—stop-and-wait and
bits at one time. IBM’s SDLC (Synchronous sliding window.
Data Link Control) and HDLC (High-level Data
Stop-and-Wait
Link Control) are examples of bit-oriented In a stop-and-wait method, the sender waits for an
protocols. Most LANs use bit-oriented framing. acknowledgement after every frame it sends. Only after
There is usually a maximum frame size. For an acknowledgement is received the next frame is sent.
example, Ethernet has a maximum frame size This process of alternately sending and waiting
of 1,526 bytes. The beginning and end of a continues until the sender transmits an end of
frame is signaled with a special bit sequence transmission (EOT) frame.
(01111110 for HDLC). If no data is being The advantage of stop-and-wait is its simplicity—
transmitted, this same sequence is continuously each frame is checked and acknowledged before the
transmitted so the end systems remain next frame is sent. The disadvantage is inefficiency—it
synchronized. is very slow. Each frame has to travel all the way to the
Clock-based Framing: In a clock-based receiver and an acknowledgement must go back all the
system, a series of repetitive pulses are used to way to sender before the next frame can be sent. Thus,
maintain a constant bit rate and keep the digital each frame is alone on the line. Each frame sent and
bits aligned in the data stream. SONET received uses the entire time needed to cross the link. If
(Synchronous Optical Network) is a the distance is long, the time spent waiting for ACKs
synchronous system in which all the clocks in between each frame adds badly to the total transmission
the network are synchronized back to a master time.
82 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Stop-and-Walt

Stop-and-Wait ARQ assumes that the last data frame was lost in
Features Added to Stop-and-Wait Flow transit and sends it again.
Control for Retransmission Work How Stop-and-Wait-ARQ Works
Stop-and-wait ARQ is a form of stop-and-wait flow To Handle Damaged Frames?
control that includes retransmission of data in case of When a frame is discovered by the receiver to
lost or damaged frames. For retransmission purpose, contain an error, it returns a NAK frame and the sender
four features have been added to the basic flow control retransmits the last frame.
mechanism: For example, in the Figure given on the next page,
The sending device keeps a copy of the last the sender transmits a data frame—data1. The receiver
frame transmitted until it receives an returns an ACK1, indicating that data 1 arrived
acknowledgement for that frame. This helps undamaged and it is now expecting data 0. The sender
the sender in retransmitting lost or damaged transmits its next fame—data 0. It arrives undamaged,
frames until they are received correctly. and the receiver returns ACK 0. The sender transmits
For identification purposes, both data frames its next frame—data1. The receiver discovers an error
and ACK frames are numbered alternately 0 in data1 and returns a NAK. The sender retransmits
and 1. A data 1 frame is acknowledged by an data1. This time data1 arrives intact, and the receiver
ACK1 frame, indicating that the receiver did returns ACK1.
receive data 1 and is now expecting data 0. How Stop-and Wait ARQ Works
This numbering helps in identification of data to Handle Lost Frames?
frames in case of duplicate transmission for There are three frame types which can be lost in
lost acknowledgements in particular. transit.
If an error is discovered in a data frame, a NAK Lost Data Frame
frame is returned. NAK frames are not Figure given on the next page shows how stop-wait
numbered. NAK tells the sender to retransmit ARQ handles the loss of a data frame. The sender is
the last frame sent. equipped with a timer that starts every time a data frame
The sending device is equipped with a timer. is transmitted. If the frame never makes it to the receiver,
If an expected acknowledgement is not the receiver can never acknowledge it, positively or
received within a set time period, the sender negatively. The sending device waits for an ACK or
 PHYSICAL AND DATA LINK LAYER / 83

NAK frame until its timer goes off, at which point it tries again. It retransmits the last data frame, restarts its timer,
and waits for an acknowledgement.
Lost Acknowledgement
In this case, the data frame has made it to the receiver but the ACK or NAK frame returned by the receiver is
lost in transit. The sending device waits until its timer goes off, then reransmits the data frame. The receiver checks
the number of the new data frame. If the lost frame was a NAK, the receiver accepts the new copy and returns the
appropriate ACK (assuming the copy arrives undamaged). If the lost frame was an ACK, the receiver recognizes
the new copy as a duplicate, acknowledges its receipt, then discards it and waits for the next frame.

Stop-and-Wait ARQ lost data frame

..
.

Stop-and-Wait ARQ lost ACK frame

84 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Stop-and-WaIt ARQ lost ACK frame

GO-BACK-N ARQ In Figure given on next page, data 0 and data 1

GO-Back-N ARQ Method of Error Correction in arrive intact but data 2 is lost. The next frame to arrive
Data Transmission with Reference to Various Frames
at the receiver is data 3. The receiver is expecting data
In this sliding window Go-back-N ARQ method, if
one frame is lost or damaged, all frames sent since the 2 and so considers data 3 to be an error, discards it, and
last frame acknowledged are retransmitted. returns a NAK2, indicating that 0 and 1 have been
Damaged Frame: Suppose frames 0, 1, 0, and 2 accepted, but 2 is in error (in this case lost). In the figure,
have been transmitted and the first acknowledgement because the sender has transmitted data 4 before
received is a NAK2. A NAK means (1) a positive receiving the NAK2, data 4 arrives at the destination
acknowledgement of all frames received prior to the out of sequence and is therefore, discarded. Once the
damaged frame, and (2) a negative acknowledgement sender receives the NAK2, it retransmits all three
of the frame indicated. If the first acknowledgement is pending frames (2, 3 and 4).
a NAK2, it means that data frames 0, and 1 were all
Lost Acknowledgement: The sending device can
received in good shape. Only frame 2 must be resent.
When frames 0 through 5 have been transmitted send as many frames as the window allows before
before a NAK is received for frame 3. It means data 0 waiting for an acknowledgement. Once that limit has
and data 1 and 2 which were received before the damaged been reached or the sender has no more frames to send,
frame, have already been accepted. The retransmission, it must wait. The sender is equipped with a timer that
therefore, consists of frames 3, 4, and 5. begins counting whenever the window capacity is
Lost Data Frame: For sliding window protocols reached. If an acknowledgement has not been received
data frames are to be transmitted sequentially. If one or within the time limit, the sender retransmits every frame
more frames are so noise corrupted that they become transmitted since the last ACK.
lost in transit, the next frame to arrive at the receiver
Figure given on next page, shows a situation in
will be out of sequence. The receiver checks the
which the sender has transmitted all of its frames and is
identifying number of each frame, discovers that one or
more have been shipped, and returns a NAK for the waiting for an acknowledgement that has been lost along
first missing frame. A NAK frame does not indicate the way. The sender waits a predetermined amount of
whether the frame has been lost or damaged, just that it time, then retransmits the unacknowledged frames. The
needs to be resent. The sending device then retransmits receiver recognizes that the new transmission is a repeat
the frame indicated by the NAK, as well as any frames of an earlier one, sends another ACK, and discards the
that it had transmitted after the lost one. duplicate data.
 PHYSICAL AND DATA LINK LAYER / 85

Go-back-n, Lost Data Frame

Go-back-n, lost ACK

Selective Repeat ARQ: In selective repeat ARQ, received after a NAK has been sent until the
only the specific damaged or lost frame is retransmitted. damaged frame has been replaced.
If a frame is corrupted in transit, a NAK is returned and The sending device contains a searching
the frame is resent out of sequence. The receiving device mechanism to find and select only the
is able to sort the frames it has and insert the corrected requested frame for retransmission.
frame into its proper place in the sequence. To make A buffer in the receiver keeps all previously
such selectivity possible, a selective repeat ARQ system received frames on hold until all retrans-
differs from a go-back-n ARQ system in the following mission have been sorted and any duplicate
ways. frames have been identified and discarded.
The receiving device contains sorting logic To aid selectivity, ACK numbers, like NAK
to enable it to reorder frame received out of numbers, refer to the frame received (or lost)
sequence. It must also be able to store frames instead of the next frame expected.
86 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

This complexity requires a smaller window Lost Frames: Although frames are accepted out
size that is needed by the go-back-n method of sequence, they are not acknowledged out of sequence.
if it is to work efficiently. The window size is If a frame is lost, the next frame will arrive out of
less than or equal to (n + 1)2, where n–1 is the sequence. When the receiver tries to reorder the existing
go-back-n window size. frames to include it, it will discover the discrepancy
Damaged Frames: Figure given on the next page and return a NAK. The receiver recognizes the omission
shows a situation in which a frame is received in error.
only if other frames follow. If the lost frame was the
Frames 0 and 1 are received but not acknowledged.
last of the transmission the receiver does nothing and
Data2 arrives and is found to contain an error, so a
NAK2 is returned. Like NAK frames in go-back-n error the sender treats the silence like a lost acknowledgement.
correction, a NAK here acknowledges both the intact Lost Acknowledgement: Lost ACK and NAK
receipt of any previously unacknowledged data frames frames are treated by selective repeat ARQ just as they
and indicates an error in the current frame. are by go-back-n ARQ. When the sending device reaches
The method used to handle error control depends either the capacity of its window or the end of its
on the method used for flow control. For stop-and-wait transmission, it sets a timer. If no acknowledgement
flow control, stop-and-wait ARQ is used. For sliding arrives in the time allotted, the sender retransmits all of
window flow control, go-back-n or selective repeat ARQ the frames that remain unacknowledged. In most cases,
is used.
the receiver recognizes any duplications and discards
In stop-and-wait ARQ the unacknowledged frame
them.
is retransmitted.
In go-back n ARQ, retransmission begins with the Pipelining: Pipelining is the process of buffering
last unacknowledged frame even it subsequent frames in line a number of tasks to be processed. It is used only
have arrived correctly. Duplicate frames are discarded. in Go-Back-N ARQ and selective repeat ARQ (and not
In selective repeat ARQ only the unacknowledged in stop and wait ARQ). This is because in both these
frame is retransmitted. methods multiple frames can be transmitted without
NAK2 informs the sender that data 0 and data 1 waiting for acknowledgement for frames sent earlier.
have been accepted, but that data 2 is to be resent. Unlike Pipelining in the network is, thus, a task that starts
in a go-back-n system, however, the receiver in a before the completion of the earlier one. It certainly
selective repeat system continues to accept new frames
improves the efficiency of bandwidth utilization. An
while waiting for an error to be corrected. In the figure,
example of pipelining the giving second order of printing
the receiver accepts data 3, 4, and 5 while waiting for a
new copy of data 2. When the new data 2 arrives, an to the prints while the first order of printing is stille
ACK5 is returned, acknowledging the new data 2 and being processed.
the original frames 3, 4, and 5.

Pipe Line in Go-Block-N

 PHYSICAL AND DATA LINK LAYER / 87

Use of pipelining in Go-Back-N ARQ.

In the as diagram, a frame 0 is received by the receiver and without waiting for receiver and without waiting for
acknowledgement of frame 0 at sender’s site, the sender is permitted to transmit next frame 1. This process is called
pipelining.

Selective Repeat Pipelining

Pipelining in Selective Repeat: Here, frame 0 and frame 1 are sent in continuous order without the sender
waiting for acknowledgement for frame sent first. Pipelining improves overall utilization of bandwidth is clearly
seen in the above figure.
Piggybacking: In piggybacking an acknowledgement of frame is appended with the data frame. The process
can be used if both sender and receiver have some data to transmit. It definitely improves the overall efficiency of
the transmission. While data is being sent by both sender and receiver they maintain control variable S and R
respectively.
In the following figure, both sender and receiver are transmitting data.

Piggybacking

We note that sender sends frame o(Fo) with ACKo appended to it. The receiver sends frame o(Fo) with ACK
1 appended to it. This method of transmitting both frame and acknowledgement together naturally increases efficiency
of bandwidth utilization because piggybacking ACKS with frames saves both time and bandwidth.
Sliding Window Protocol: (a) Initial level (b) first frame sent (c) Frame sent and piggybacked the ACK (d)
Window size is set to 1 as frame acknowledged.
88 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

The third window indicates the frame sent and

piggybacked the acknowledgement. The last window
of the receiver side indicates the received
acknowledgement and window size is set to 1.
MULTI-ACCESS COMMUNICATION
The IEEE has sub-divided the data link layer into
two sub-layers and logical link control and medium
access control (MAC).
The medium access control sublayer, resolves the
contention for the shared media. It contains the
The sliding window protocol remains synchronized synchronization flag flow and error control specification
necessary to move information from one place to
in any condition either garbled frames, lost frames or
another, as well as the physical address of the next station
timeout frames. In the sliding window method, the
to receive and route a packet MAC protocols are specific
sender can transmit several frames before requiring the
to the LAN using them Ethernet token Ring, and Token
acknowledgement. The term window in sliding window Bus etc.
refers to an extra buffer created by both sender and the CSMA/CD
receiver. The main idea behind the sliding window CSMA is a network access method used on shared
protocol is to assign the sequence number to every network topologies such as Ethernet to control access
outgoing frame starting from 0 to 2n–1. Here, n is the to the network. Devices attached to the network cable
size of the window and window size indicates the listen (carrier sense) before transmitting. If the channel
maximum amount of data that can be sent before an is in use, devices wait before transmitting. MA (multiple
access) indicates that many devices can connect to and
acknowledgement arrives The link can carry multiple
share the same network. All devices have equal access
frames at once and the capacity can be used efficiently. to use the network when it is clear. Even though devices
The frames are numbered from 0 to n – 1. attempt to sense whether the network is in use, there is
For example n = 8 .The frames are numbered 0, 1, a good chance that two stations will attempt to access it
2, 3, 4, 5, 6, 7. at the same time. On large networks, the transmission
According to the diagram, the sender starts with time between one end of the cable and another is enough
the data to be sent, fills the window with the data and that one station may access the cable even though
transmits the copies. The receiver must have the buffer another has already just accessed it. There are two
space ready to receive the packet arrived. When the methods for avoiding these so-called collisions, listed
packet arrives in the sequence the receiver passes the here:
packet to the receiving application and transmits the CSMA/CD (carrier sense multiple access/
acknowledgement to the sender. When the receiver sends collision detection). CD (collision detection) defines
the ACK, it also sends the number of next frame it what happens when two devices sense a clear channel,
expects to receive. The sequence number with in the then attempt to transmit at the same time. A collision
sender’s window indicates frames sent but not yet occurs, and both devices stop transmission, wait for a
acknowledgement. The acknowledgement is either sent random amount of time, then retransmit. This is the
technique used to access the 802.3 Ethernet network
in some special control frame or use the concept of
channel. This method handles collisions as they occur,
Piggybacking with data frames in the reverse direction.
but if the bus is constantly busy, collisions can occur so
According to the diagram of the sender’s window.
often that performance drops drastically. It is estimated
The first part is the initial level when data is sent to that network traffic must be less than 40 per cent of the
the window. The second window indicates the first frame bus capacity for the network to operate efficiently. If
sent. The third window indicates the frame sent but not distances are long, time lags occur that may result in
yet acknowledgement. The forth window indicates the inappropriate carrier sensing, and hence collisions.
window size is set to 1 as frame acknowledgement. CSMA/CA (carrier sense multiple access/
The Receiver’s Window collision avoidance) In CA (collision avoidance),
The first window indicates the receiving of the collisions are avoided because each node signals its
frame. The second window indicates the first frame sent. intent to transmit before actually doing so. This method
 PHYSICAL AND DATA LINK LAYER / 89

is not popular because it requires excessive overhead limited to roughly 100 metres, but Ethernet networks
that reduces performance. can be easily extended to link entire schools or office
ETHERNET buildings using network bridge devices.
Ethernet is a physical and data link layer technology IEEE 802.3 defines two categories: baseband and
for local area networks (LANs). Ethernet was invented broadband
by engineer Robert Metcalfe.When first widely
deployed in the 1980s, Ethernet supported a maximum
theoretical data rate of 10 megabits per second (Mbps).
Later, so-called “Fast Ethernet” standards increased this
maximum data rate to 100 Mbps. Today, Gigabit
Ethernet technology further extends peak performance
up to 1000 Mbps.
Higher level network protocols like Internet Addressing: Each station on an ethernet network
Protocol (IP) use Ethernet as their transmission medium. has its own network Interface card. The NIC usually
Data travels over Ethernet inside protocol units called
fits inside the station and provides the station with a six
frames.The run length of individual Ethernet cables is
byte physical address. The number on the NIC is unique.

(i) Preamble: The preamble contains seven bytes Reverse Address Resolution Protocol (RARP). ARP is
of alternative 0s and 1s that alert the receiving system used to resolve the ethernet address of a NIC from an
to the coming frame and enable it to synchronize its IP address in order to construct an ethernet packet
input timing. around an IP data packet. This must happen in order to
(ii) Start Frame Delimiter (SFD): The SFD tells send any data across the network. Reverse address
the receiver that every thing that follows is data, starting resolution protocol (RARP) is used for diskless
with the addresses. computers to determine their IP address using the
(iii) Destination Address (DA): The destination network.
address field is allotted six bytes and contains the Address Resolution Protocol (ARP): In an earlier
physical address of the packet next destination. section, there was an example where a chat programme
(iv) Source Address: The source address field is was written to communicate between two servers. To
also allotted six bytes and contains the physical address send data, the user (Tom) would type text into a dialog
of the last device to forward the packet. box, hit send and the following happened:
(v) PDU: These next 2 bytes indicate the no. of 1. The programme passed Tom’s typed text in a
bytes is coming PDU. If the length of the PDU is fixed, buffer, to the socket.
this field can be used to indicate type or as a base for 2. The data was put inside a TCP data packet with
other protocols. a TCP header added to the data. This header
(vi) Data and Padding: The PDU can be anywhere contained a source and destination port number
from 46 to 1500 bytes long depending on the type of along with some other information and a
frame and the length of the information field. The PDU checksum.
is generated by the upper sublayer then linked to the 3. The TCP packet was be placed inside an IP data
802.3 frame. packet with a source and destination IP address
(vii) CRC: The last field in the 802.3 frame contains
along with some other data for network
the error detection information in this case a CRC-32.
management.
ARP AND RARP
4. The IP data packet was placed inside an ethernet
Address Resolution Protocol (ARP) provides a
data packet. This data packet includes the
completely different function to the network than
90 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

destination and source address of the network 4. Byte size of the protocol address (1 byte). 4
interface cards (NIC) on the two computers. The 5. Type of operation. 1 = ARP request, 2 = ARP
address here is the hardware address of the reply, 3 = RARP request, 4 = RARP reply.
respective cards and is called the MAC address. 6. The sender’s ethernet address (6 bytes)
5. The ethernet packet was transmitted over the 7. The sender’s IP address (4 bytes)
network line. 8. The recipient’s ethernet address (6 bytes)
9. The recipient’s IP address (4 bytes)
6. With a direct connection between the two
When the ARP reply is sent, the recipient’s ethernet
computers, the network interface card on the
address is left blank.
intended machine, recognized its address and
In order to increase the efficiency of the network
grabbed the data.
and not tie up bandwidth doing ARP broadcasting, each
7. The IP data packet was extracted from the
computer keeps a table of IP addresses and matching
ethernet data packet.
ethernet addresses in memory. This is called ARP cache.
8. The TCP data packet was extracted from the IP Before sending a broadcast, the sending computer will
data packet. check to see if the information is in it’s ARP cache. If it
9. The data was extracted from the TCP packet is it will complete the ethernet data packet without an
and the program displayed the retrieved data ARP broadcast. Each entry normally lasts 20 minutes
(text) in the text display window for the intended after it is created. RFC 1122 specifies that it should be
recipient to read. possible to configure the ARP cache timeout value on
In step 4 above, the IP data was going to be placed the host. To examine the cache on a Windows, UNIX,
inside an ethernet data packet, but the computer or Linux computer type “arp -a”.
constructing the packet does not have the ethernet If the receiving host is on another network, the
address of the recipient’s computer. The computer that sending computer will go through its route table and
is sending the data, in order to create the ethernet part determine the correct router (A router should be between
of the packet, must get the ethernet hardware (MAC) two or more networks) to send to, and it will substitute
address of the computer with the intended IP address. the ethernet address of the router in the ethernet message.
This must be accomplished before the ethernet packet The encased IP address will still have the intended IP
can be constructed. The ethernet device driver software address. When the router gets the message, it looks at
on the receiving computer is not programmed to look at the IP data to tell where to send the data next. If the
IP addresses encased in the ethernet packet. If it did, recipient is on a network the router is connected to, it
the protocols could not be independent and changes to will do the ARP resolution either using it’s ARP buffer
one would affect the other. This is where address cache or broadcasting.
resolution protocol (ARP) is used. Tom’s computer Reverse Address Resolution Protocol (RARP):
sends a network broadcast asking the computer that has As mentioned earlier, reverse address resolution
the recipient’s IP address to send it’s ethernet address. protocol (RARP) is used for diskless computers to
This is done by broadcasting. The ethernet destination determine their IP address using the network. The RARP
is set with all bits on so all ethernet cards on the network message format is very similar to the ARP format. When
will receive the data packet. The ARP message consists the booting computer sends the broadcast ARP request,
of an ethernet header and ARP packet. The ethernet it places its own hardware address in both the sending
header contains: and receiving fields in the encapsulated ARP data packet.
1. A 6 byte ethernet destination address. The RARP server will fill in the correct sending and
2. A 6 byte ethernet source address. receiving IP addresses in its response to the message.
3. A 2 byte frame type. The frame type is 0806 This way the booting computer will know its IP address
hexadecimal for ARP and 8035 for RARP when it gets the message from the RARP server.
The encapsulated ARP data packet contains the
following: Q. 1. Write down the characteristics of CDMA.
1. Type of hardware address (2 bytes). 1= ethernet. Ans. 1. In CDMA, each user is given a unique code
2. Type of protocol address being mapped (2 sequence or signature sequence. This sequence allows
bytes). 0800H (hexadecimal) = IP address. the user to spread the information signal across the
3. Byte size of the hardware address (1 byte). 6 assigned frequency band.
 PHYSICAL AND DATA LINK LAYER / 91

2. In CDMA, the users access the channel in random Q. 7. How the receiver will detect that there is
manner. an error in:
3. CDMA signals are spread in frequency. Therefore 11101110 11011110 11100100 1101100011001001?
demodulation? Ans. The receiver counts the 1’s in each character
Q. 2. What is an Error? and comes up with even numbers (b, 6, 4, 4, 4). The
Ans. An error is the change or the mismatching data are accepted.
take place between the data unit sent by transmitter and Q. 8. Suppose the word HELLO is corrupted
the data unit received by the receiver e.g. 10101010 during transmission?
sent by sender 10101011 received by receiver. Here, is 11111110 11011110 11101100 110111000 11001001?
an error of 1 bit. How receiver will check it out?
Q. 3. Define Error Control. Ans. The receiver counts the 1’s in each character
Ans. Error control refers to mechanisms to detect and comes up with even and odd numbers (7, 6, 5, 4, 4).
and correct errors that occur in the transmission of The receiver knows that the data are corrupted, discards
frames. The most common techniques for error control them and asks for Retransmission.
are based on some or all of the following: Q. 9. Explain about error correction.
1. Error detection Ans. Error correction is the mechanism by which
2. Positive acknowledgement we can make changes in the received erroneous data to
3. Retransmission after time-out make it free from error.
4. Negative acknowledgement and retransmission. The two most common error correction mechanisms
These mechanisms are also referred as automatic are:
repeat request (ARC)). (i) Error correction by Retransmission.
Q. 4. How can the simple parity bit detect a (ii) Forward Error Correction.
damaged data unit? Q. 10. What is check sum?
Ans. In this technique, a redundant bit called a Ans. Checksum is the one of the method used for
parity bit, is added to every data unit so that the total error detection, based on the concept of redundancy. In
number of Is in the unit becomes even (or odd). Suppose this mechanism, the unit is divided into K sections, each
we want to transmit 1100001. Adding the number of of n bits. All sections are added using ones complement
1’s gives us 3, an odd number. Before transmitting, we to get the sum. This is complemented and becomes the
pass the data unit through a parity generator. The parity check sum. There after this check sum is sent with the
generator counts the 1’s and appends the parity bit to data. At the receiver side the unit is divided into K
the end (all in this case). sections each of n bits. All sections are added using ones
Q. 5. What is the difference between even parity complement to get the sum. The sum is complemented.
and odd parity? If the result is zero data are accepted otherwise rejected.
Ans. In case of redundancy check method we have Q. 11. What are the different types of error?
to append the data unit with some extra bits. These extra Or
bits are called parity.
How a single bit error does differ from a burst
This parity or parity hit can be even or odd.
In case of even parity we have to make number of error?
1’s even, including the parity hit e.g. 1110001 is the Ans. A single bit error is an isolated error condition
data unit where the no. of l’s is already even then we that alters one bit but does not affect nearby bits. On
will insert 0 at the next to data unit it’, 1110001. In case the other hand A burst error is a contiguous sequence of
of odd parity we have to make no. of l’s odd, including bits in which the first and last bits and any number of
the parity bit. e.g. 1111000 is the data unit, where the
no. of 1’s is even then we will insert I at the next to data intermediate bits are received in error.
unit i.e. 11110001. A single bit can occur in the preserve of while noise,
Q. 6. Suppose the sender wants the word when a slight random deterioration of single-to-noise
“HELLO”. In ASCII the five characters are coded ratio is sufficient to confuse the receiver’s decision of a
as: single bit. On the other hand burst errors are more
1110111 1101111 1110010 1101100 1100100
common and more difficult to deal with. Burst error
H E L L O
What will be the combination of actual bits to can be caused by impulse noise.
send? Q. 12. Explain any one mechanism used for error
Ans. 11101110 11011110 11100100 11011000 detection?
11001001
92 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Or data unit. Then we organize them into table. Shows in

What is the parity check method of error fig we have four data units shown in four rows and eight
detection? columns. We then calculate the parity hit for each column
Ans. The most common and least expensive and create a new row of 8 bits. They are the parity bits
mechanism for error detection is the parity check. for the whole block. The first parity bit in the fifth row
Parity checking can be simple or two-dimensional. is calculated based on all first bits, the second parity bit
is calculated based on all second bits, and so on. We
then attach the 8 parity bits to the original data and sent
them to the receiver.
Simple Parity Check: In this technique, a
redundant bit, called a parity bit, is added to every data
unit so that the total number of Is in the unit (including
the parity bit) becomes even (or odd). Suppose we want
to transmit the binary data unit 1100001
Receiver Node Sender Node

Q. 13. Explain CRC method of error detection?

Ans. Cyclic Redundancy Check (CRC): Cyclic
Redundancy check method is most powerful mechanism
of error detecting. Unlike the parity check which is based
on addition, CRC is based on binary division.
Transmission Mode: Adding the no. of is giving In CRC, instead of adding bits to achieve a desired
us 3 an odd number. Before transmitting we pass the parity, a sequence of redundant bits, called the CRC or
data unit through a parity generator. The parity generator the CRC remainder, is appended to the end of a data
counts the is and appends the parity bit to the end. The unit so that the resulting data unit becomes exactly
total no. of is now 4, an even number. The system now divisible by a second predetermined binary number. At
transmits the entire expanded unit across the network its destination the incoming data unit is divided by the
link. When it reaches its destination, the receiver puts same number. If at this step there is no remainder, the
all 8 bits through an even parity checking function. If data unit is assumed to be intact and is therefore
the receiver sees 11000011, it counts four is, an even accepted. A remainder indicates that the data unit has
number and the data unit passes. But, if instead of been damaged in transit and therefore must be rejected.
11000011, the receiver sees 11001011 then when the The redundancy bits used by CRC are derived by
parity checker counts the Is it gets 5 an odd number. dividing the data unit by a predetermined divisor, the
remainder is the CRC. A CRC must have two qualities.
The receiver knows that an error has been introduced
It must have exactly one less bit than the divisor, and
into the data somewhere and therefore rejects the whole
appending it to the end of the data string must make the
unit.
resulting bit sequence exactly divisible by the divisor.
Two Dimensional Parity Check: A better CRC generator and checker
approach is the two dimensional parity check in this First, a string of n 0’s is appended to the data unit.
method, a block of bits is organized in a table (rows The number n is less than the number of bits in the
and columns). First we calculate the parity bit for each predetermined divisor, which are n + 1 bits.
 PHYSICAL AND DATA LINK LAYER / 93

Second, the newly formed data unit is divided by the string has been changed in transit the division yields
the divisor, using a process called binary division the a non zero remainder and the data unit does not pass.
remainder resulting from this division is the CRC. Q. 14. How the data communication between
Third, the CRC of n bits derived in step 2 replaces sender and the receiver will take place where the
the appended Os at the end of the data unit. The data error detection method is checksum and the data is:
 10101001 00111001
unit arrives at the receiver data first followed by the Ans. Sender: The numbers are added using one’s
CRC. The receiver treats the whole string as a unit and complement arithmetic
divides it by the same divisor that was used to find the 10101001
CRC remainder. 00111001
Sum 11100010
Checksum 00011101 (I’s Complement)
the pattern unit is
 10101001 00111001 00011101
Checksum
Receiver
10101001 00111001 00011101
When the receiver adds the three sections, it will
get all is, which after complementing is all 0’s and shows
that there is no error.
10101001
Receiver Sender 00111001
00011101
If the string arrives without error, the CRC checker
yields a remainder of zero and the data unit passes. If Sum 11111111
Complement 00000000 means that the pattern is ok.
Q. 15. What is Hamming Code of Error Correction? How it calculate, the redundancy?
Or
Explain any one method used for error correction.
Ans. The Hamming Code can be applied to data units of any length and uses the relationship between data and
redundancy bits.
Suppose there are 7 bits ASCII codes which requires 4 redundancy bits that can be added to the end of the data
unit or interspersed with the original data bits. These units are position in 1, 2, 4, arid 8 (the position is in an 11 bit
sequence that are powers of 2). We prefer these bits are r1, r2, r4 and r8.
11 10 9 8 7 6 5 4 3 2 1
d d d r8 d d d r4 d r2 r1
In the hamming code, each r bit is the parity bit for the combination of data bits as shown below:
r1 bits 1, 3, 5, 7, 9, 11
r2 bits 2, 3, 6, 7, 10, 11
r4 bits 4, 5, 6, 7
r8 bits 8, 9, 10, 11
Each data bit may be included in more than one calculation. In the sequences e.g. each of the original data bits
is included in at least two bits, while the r bits are induced in only one.
11 9 7 5 3 1
d d d r8 d d d r4 d r2 r1
11 10 7 6 3 2
d d d r8 d d d r4 d r2 r1
7 6 3 2
94 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

d d d r8 d d d r4 d r2 r1
11 10 9 8
d d d r8 d d d r4 d r2 r1

Q. 16. What are various error correction codes? 5 4 9

Ans. A mechanism that can handle correction of 6 4 10
an error heading of error correction code categories
under the some techniques. 7 4 11
There are two methods for error correction: There is the practical solution for this method that
(1) Error correction by retransmission. is “Hamming Code”.
(2) Forward error correction. Q. 17. What is the difference between ARP and
Error Correction by Retransmission: In error RARP?
correction by retransmission, when an error is Ans. Address Resolution Protocol (ARP) provides
discovered, the receiver can have the sender retransmit a completely different function to the network than
the entire data unit.
Reverse Address Resolution Protocol (RARP). ARP is
Forward Error Correction: In forward error
used to resolve the Ethernet address of a NIC from an
correction (FEC), a receiver can use an error-correcting
IP address in order to construct an Ethernet packet
mode, which automatically corrects certain errors. In
theory it is possible to correct any error automatically. around an IP data packet. This must happen in order to
Error correcting codes however, are more sophisticated send any data across the network. Reverse address
than error detection codes and require more redundancy resolution protocol (RARP) is used for disk less
bits. computers to determine their IP address using the
e.g. To correct a single bit error in an ASCII network.
character, the error correction code must determine Q. 18. What is the difference between
which of the 7 bits has changed, In this case we have to connectionless and connection-oriented protocol?
distinguish between eight different states no error, error Ans. 1. In connectionless communication there is
in position 2, and so on, up to the error in position 7. To no need to establish connection between source (sender)
do so requires enough bits to show all eight states. and destination (receiver). But in connection-oriented
At first glance, it seems that or 3-bit redundancy communication connection must established before data
code should be adequate because 3 bits can show eight transfer.
different states (000 to 111) and can therefore indicate 2. Connection-oriented communication is more
the locations of eight different possibilities. To calculate reliable than connectionless communication.
the no. of redundancy bits. We should consider 3. In connection-oriented communication infor-
2r > = m + r + 1
mation can be resend if there is an error in receiver side
Where m is the no. of bits to be transfer r stands for
the no. of redundancy. By this manner. ( missing data, corrupt data etc. ).But in connectionless
communication it is not possible because the destination
Number of Number of Total Bits
data Bits m Redundancy Bits r m+r does not inform the source about data is received or
not.
1 2 3
4. Connection-oriented communication have higher
2 3 5 overhead and and place greater demands on bandwidth.
3 3 6 But in connectionless communication requires far less
4 3 7 overhead than connection-oriented communication.
 95

Internetworking Devices
8
An Ethernet network interface card is installed in
Two or more devices connected for the purpose of an available slot inside the computer. The NIC assigns
sharing data or resources can form a network. Putting a unique address called a MAC (media access control)
together a network is often more complicated than to the machine. The MACs on the network are used to
simply plugging cable into a hub. A device called a direct traffic between the computers. The back plate of
repeater or regenerator is inserted into the network to thenetwork interface card features a port that looks
increase the coverable distance. In the second a device similar to a phone jack, but is slightly larger. This port
called a bridge is inserted for traffic management. accommodates an Ethernet cable, which resembles a
When two or more separate networks are connected thicker version of a standard telephone line. Ethernet
for exchanging data or resources they become an cable must run from each network interface card to a
internetwork. Linking a number of LANs into an internet central hub or switch. The hub or switch acts like a relay,
requires additional internetworking devices called passing information between computers using the MAC
routers and gateways. These devices are designed to addresses and allowing resources like printers and
overcome obstacles to interconnection without scanners to be shared along with data.
disrupting the independent functining of the networks. A network interface card does not have to be hard
wired with physical cable. Wireless Ethernet cards are
installed like their wired counterparts, but rather than a
NETWORK INTERFACE CARDS port for an Ethernet cable, the card features a small
A network interface controller (also known as a antenna. The card communicates with the central
network interface card, network adapter, LAN adapter wireless switch or hub via radio waves. Wireless LANs
and by similar terms) is a computer hardware component may have some restrictions depending on the material
that connects a computer to a computer network. the building is made from. For example, lead in walls
A network interface card, more commonly referred can block signals between the network interface card
to as a NIC, is a device that allows computers to be and hub or switch.
joined together in a LAN, or local area network. When buying components for a LAN, make sure
Networked computers communicate with each other the NICs and hub or switch have the same capabilities.
using a given protocol or agreed-upon language for The entire network must be either wired or wireless, so
transmitting data packets between the different a wireless networkinterface card cannot talk to a wired
machines, known as nodes. The network interface card switch or hub. In addition, newer versions of hardware
acts as the liaison for the machine to both send and will likely support more features and/or greater speeds
receive data on the LAN. than older versions. Make sure your central switch or
The most common language or protocol for LANs hub can utilize the highest capabilities of the network
is Ethernet, sometimes referred to as IEEE 802.3. A interface card under consideration.
lesser-used protocol is Token Ring. When building a For those who wish to connect LANs located in
LAN, a network interface card must be installed in each different areas of the city, state or country, ATM
computer on the network and all NICs in the network (asynchronous transfer mode) can create wide area
must be of the same architecture. For example, all must networks or WANs by connecting LANs together. LANs
either be Ethernet cards, Token Ring cards, or an are still built with a network interface card in each
alternate technology. networked computer, but ATM uses broadband Internet
96 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

access to link the LANs to online ATM switches. This Because repeaters work with the actual physical
type of ATM WAN is referred to as an Internetwork. signal, and do not attempt to interpret the data being
MODEMS transmitted, they operate on the Physical layer, the first
A modem (modulator-demodulator) is a device that layer of the OSI model.
modulates an analog carrier signal to encode digital A “digipeater” is a blend meaning “digital repeater”,
information, and also demodulates such a carrier signal particularly used in amateur radio. Store and forward
to decode the transmitted information. The goal is to digipeaters generally receive apacket radio transmission
produce a signal that can be transmitted easily and and then retransmit it on the same frequency, unlike
decoded to reproduce the original digital data. Modems repeaters that receive on one and transmit on another
can be used over any means of transmitting analog frequency.
signals, from light emitting diodes to radio. The most A repeater connects two segments of your network
familiar example is a voice band modem that turns the cable. It retimes and regenerates the signals to proper
digital data of a personal computer into modulated amplitudes and sends them to the other segments. When
electrical signals in the voice frequency range of a talking about, ethernet topology, you are probably
telephone channel. These signals can be transmitted over talking about using a hub as a repeater. Repeaters require
telephone lines and demodulated by another modem at a small amount of time to regenerate the signal. This
the receiver side to recover the digital data. can cause a propagation delay which can affect network
Modems are generally classified by the amount of communication when there are several repeaters in a
data they can send in a given unit of time, usually row. Many network architectures limit the number of
expressed in bits per second (bit/s, or bps). Modems repeaters that can be used in a row. Repeaters work only
can alternatively be classified by their symbol rate, at the physical layer of the OSI network model.
measured in baud. The baud unit denotes symbols per HUBS
second, or the number of times per second the modem A hub includes a series of ports that each accept a
sends a new signal. For example, the ITU V.21 standard network cable. Small hubs network four computers.
used audio frequency-shift keying, that is to say, tones They contain four or sometimes five ports, the fifth port
of different frequencies, with two possible frequencies being reserved for “uplink” connections to another hub
corresponding to two distinct symbols (or one bit per or similar device. Larger hubs contain eight, 12, 16,
symbol), to carry 300 bits per second using 300 baud. and even 24 ports.
By contrast, the original ITU V.22 standard, which was Key Features of Hubs: Hubs classify as Layer 1
able to transmit and receive four distinct symbols (two devices in the OSI model. At the physical layer, hubs
bits per symbol), handled 1,200 bit/s by sending 600 can support little in the way of sophisticated networking.
symbols per second (600 baud) using phase shift keying. Hubs do not read any of the data passing through them
REPEATERS and are not aware of their source or destination.
A repeater is an electronic device that receives a Essentially, a hub simply receives incomingpackets,
signal and retransmits it at a higher level and/or higher possibly amplifies the electrical signal, and broadcasts
power, or onto the other side of an obstruction, so that these packets out to all devices on the network–including
the signal can cover longer distances. the one that originally sent the packet!
The term “repeater” originated with telegraphy and Technically speaking, three different types of hubs
referred to an electromechanical device used by the army exist:
to regenerate telegraph signals. Use of the term has passive
continued in telephony and data communications. active
In telecommunication, the term repeater has the intelligent
following standardized meanings: Passive hubs do not amplify the electrical signal of
1. An analog device that amplifies an input signal incoming packets before broadcasting them out to the
regardless of its nature (analog or digital). network. Active hubs, on the other hand, do perform
2. A digital device that amplifies, reshapes, this amplification, as does a different type of dedicated
retimes, or performs a combination of any of network device called a repeater. Some people use the
these functions on a digital input signal for termsconcentrator when referring to a passive hub and
retransmission. multiport repeater when referring to an active hub.
 INTERNETWORKING DEVICES / 97

Intelligent hubs add extra features to an active hub in a home or office. This is a trivial case of bridging, in
that are of particular importance to businesses. An which the bridge learns the MAC address of each
intelligent hub typically is stackable (built in such a way connected device.
that multiple units can be placed one on top of the other Single bridges also can provide extremely high
to conserve space). It also typically includes remote performance in specialized applications such as storage
management capabilities via SNMP and virtual LAN area networks.
(VLAN) support. Classic bridges may also interconnect using a
Hubs remain a very popular device for small spanning tree protocol that disables links so that the
networks because of their low cost. A good five-port resulting local area network is a treewithout loops. In
Ethernet hub can be purchased for less than $30 USD. contrast to routers, spanning tree bridges must have
USB hubs cost only a bit more. topologies with only one active path between two points.
BRIDGES The older IEEE 802.1D spanning tree protocol could
A bridge reads the outermost section of data on the be quite slow, with forwarding stopping for 30 seconds
data packet, to tell where the message is going. It reduces while the spanning tree would reconverge. A Rapid
the traffic on other network segments, since it does not Spanning Tree Protocol was introduced as IEEE 802.1w,
send all packets. Bridges can be programmed to reject but the newest edition of IEEE 802.1D adopts the
packets from particular networks. Bridging occurs at 802.1w extensions as the base standard.
the data link layer of the OSI model, which means the The IETF is specifying the TRILL protocol, which
bridge cannot read IP addresses, but only the outermost is the application of link-state routing technology to the
hardware address of the packet. In our case the bridge layer-2 bridging problem. Devices which implement
can read the ethernet data which gives the hardware TRILL, called RBridges, combine the best features of
address of the destination address, not the IP address. both routers and bridges.
Bridges forward all broadcast messages. Only a special While “layer 2 switch” remains more of a marketing
bridge called a translation bridge will allow two
term than a technical term, the products that were
networks of different architectures to be connected.
introduced as “switches” tended to use microse-
Bridges do not normally allow connection of networks
gmentation and Full duplex to prevent collisions among
with different architectures. The hardware address is
devices connected to Ethernet. By using an
also called the MAC (media access control) address.
internalforwarding plane much faster than any interface,
To determine the network segment a MAC address
they give the impression of simultaneous paths among
belongs to, bridges use one of:
Transparent Bridging: They build a table of multiple devices.
addresses (bridging table) as they receive Once a bridge learns the topology through a
packets. If the address is not in the bridging spanning tree protocol, it forwards data link layer frames
table, the packet is forwarded to all segments using a layer 2 forwarding method. There are four
other than the one it came from. This type of forwarding methods a bridge can use, of which the
bridge is used on ethernet networks. second through fourth method were performance-
Source Route Bridging: The source computer increasing methods when used on “switch” products
provides path information inside the packet. with the same input and output port bandwidths:
This is used on Token Ring networks. 1. Store and forward: The switch buffers and
SWITCH (L2 AND L3 DIFFERENCES) verifies each frame before forwarding it.
A network switch or switching hub is a computer 2. Cut through: The switch reads only up to the
networking device that connects network segments. frame’s hardware address before starting to
The term commonly refers to a multi-port network forward it. Cut-through switches have to fall
bridge that processes and routes data at the data link back to store and forward if the outgoing port
layer (layer 2) of the OSI model. Switches that is busy at the time the packet arrives. There is
additionally process data at the network layer (Layer 3) no error checking with this method.
and above are often referred to as Layer 3 switches or 3. Fragment free: A method that attempts to retain
multilayer switches. the benefits of both store and forward and cut
Layer 2: A network bridge, operating at the data through. Fragment free checks the first 64 bytes
link layer, may interconnect a small number of devices of the frame, where addressing information is
98 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

stored. According to Ethernet specifications, While there are specialized applications, such as
collisions should be detected during the first 64 storage area networks, where the input and output
bytes of the frame, so frames that are in error interfaces are the same bandwidth, this is rarely the case
because of a collision will not be forwarded. in general LAN applications. In LANs, a switch used
This way the frame will always reach its for end user access typically concentrates lower
intended destination. Error checking of the bandwidth (e.g. 10/100 Mbit/s) into a higher bandwidth
actual data in the packet is left for the end device. (at least 1 Gbit/s). Alternatively, a switch that provides
4. Adaptive switching: A method of automatically access to server ports usually connects to them at a much
selecting between the other three modes. higher bandwidth than is used by end user devices.

Layer 3: Within the confines of the Ethernet physical layer, a layer 3 switch can perform some or all of the
functions normally performed by a router. The most common layer-3 capability is awareness of IP multicast through
IGMP snooping. With this awareness, a layer-3 switch can increase efficiency by delivering the traffic of a multicast
group only to ports where the attached device has signaled that it wants to listen to that group.

HQ Branch

ISP
 INTERNETWORKING DEVICES / 99

GATEWAYS local network to the Internet, although internal gateways

A network gateway is an internetworking system for local networks also exist.
capable of joining together two networks that use Internet default gateways are typically one of two
different base protocols. A network gateway can be types:
implemented completely in software, completely in On home or small business networks with a
hardware, or as a combination of both. Depending on broadband router to share the Internet
the types of protocols they support, network gateways connection, the home router serves as the default
can operate at any level of the OSI model. gateway.
Because a network gateway, by definition, appears On home or small business networks without a
at the edge of a network, related capabilities like router, such as for residences with dialup
Internet access, a router at the Internet Service
firewalls tend to be integrated with it. On home
Provider location serves as the default gateway.
networks, a broadband router typically serves as the
Default network gateways can also be configured
network gateway although ordinary computers can also using an ordinary computer instead of a router. These
be configured to perform equivalent functions. gateways use two network adapters, one connected to
In computer networking, a default gateway is the the local subnet and one to the outside network. Either
device that passes traffic from the local subnet to devices routers or gateway computers can be used to network
on other subnets. The default gateway often connects a local subnets such as those in larger businesses.

Internet
Gateway
HotSpot
Gateway

Media Dependent Adapter: A MDA is a plug-in

Q. 1. What are the different types of networking module allowing selection among fiber-optic, twisted
or internetworking devices? pair, and coaxial cable.
Ans. Bridge-Network segments that typically use Media Filter: When the electrical characteristics
the same communication protocol use bridges to pass of various networks are different, media filter adapter
information from one network segment to the other. connectors make the connections possible.
Gateway: When different communications Multistation Access Unit: MAUs are special
protocols are used by networks, gateways are used to concentrators or hubs for use in Token Ring networks
convert the data from the sender’s to receives. instead of Ethernet networks.
Hub: Another name for a hub is a concentrator. Network Interface Card: NICs are printed circuit
Hubs reside in the core of the LAN cabling system. The boards that are installed in computer workstations. They
hub connects workstations and sends every transmission provide the physical connection and circuitry required
to all the connected workstations. to access the network.
100 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Repeater: Connectivity device used to regenerate and amplify weak signals, thus extending the length of the
network. Repeaters perform no other action on the data.
Router: Links two or more networks together, such as an Internet Protocol network. A router receives packets
and selects the optimum path to forward the packets to other networks.
Switch: A connection device in a network that functions much like a bridge, but directs transmissions to
specific workstations rather than forwarding data to all workstations on the network.
Transceiver: The name transceiver is derived from the combination of the words transmitter and receiver. It
is a device that both transmits and receives signals and connects a computer to the network. A transceiver may be
external or located internally on the NIC.
Q. 2. What is the difference between an Intranet and the Internet?
Ans. The terms Internet, Intranet are most widely used terms. But they have a lot of difference between them.
Internet Intranet
Internet is network of computers which is open for Intranet is network of computers designed for
all. a specific group of users.
Internet itself contains a large number of intranets. Intranet can be accessed from Internet but with
restrictions.
Unlimited number of users. Limited number of users.
Visitors traffic is unlimited. Limited visitors traffic.
Contains unlimited source of information. Contains only specific group purpose
information.
Collection of various LANs, WANs and MANs. Mostly any of LAN or MAN or WAN.

Q. 3. What are L2 and L3 switches in L3 (Layer 3) switches, also called routing or IP

networking? What is the basic difference between switches, perform both switching and routing tasks.
them and how they help the networking? Because they make heavy use of integrated circuits,
Ans. L2 (Layer 2) switches operate at the second, specialized crossbar switching fabric, fast CPUs and
or data link, layer of the Open Systems Interconnection
other high-end components, they are expensive but well-
model. Their function is to check incoming network
suited to network backbone service. But less expensive
traffic, keep track of the physical addresses of all stations
attached to their ports and forward traffic to specific Layer 2 switches also provide an extensive range of
station addresses according to the forwarding tables they features and options, making them cost-effective for
have built. Layer 2 switches can deliver data at wire managing traffic at the work group and small network
speeds because they don’t have to examine individual levels.
data packets.
 101

NETWORK, TRANSPORT AND APPLICATION LAYER

Network Layer
9
packet queues may cause varying packet transfer delay.
Whenever we have multiple devices, we have Each circuit cannot be used by other callers until the
problem of how to connect them to make one-on-one circuit is released and a new connection is set up. Even
communication possible. One solution is to install a if no actual communication is taking place, the channel
point-to-point connection between each pair of devices remains unavailable to other users. Channels that are
or between a central device and every other device. available for new calls are said to be idle.
These methods however, are impractical and wasteful Virtual circuit switching is a packet switching
when applied to very large networks. A better solution technology that emulates circuit switching, in the sense
is switching. A switching network consists of a series of that the connection is established before any packets
interlinked nodes, called switches. Switches are are transferred, and packets are delivered in order.
hardware and/or software devices capable of creating While circuit switching is commonly used for
temporary connections between two or more devices connecting voice circuits, the concept of a dedicated
linked to the switch but not to each other. In a switched path persisting between two communicating parties or
network some of these nodes are connected to the nodes can be extended to signal content other than voice.
communicating devices. Others are used only for Its advantage is that it provides for continuous transfer
routing. without the overhead associated with packets making
maximal use of available bandwidth for that
communication. The disadvantage is inflexibility; the
CIRCUIT AND PACKET SWITCHING connection and the bandwidth associated with it are
Circuit Switching: It is a methodology of imple- reserved and unavailable for other uses.
menting a telecommunications network in which two Packet Switching: It is a digital networking
network nodes establish a dedicated communications communications method that groups all transmitted data
channel (circuit) through the network before the nodes regardless of content, type, or structure into suitably
may communicate. The circuit guarantees the full sized blocks, called packets. Packet switching features
bandwidth of the channel and remains connected for delivery of variable-bit-rate data streams (sequences of
the duration of the communication session. The circuit packets) over a shared network. When traversing
functions as if the nodes were physically connected as network adapters, switches, routers and other network
with an electrical circuit. nodes, packets are buffered and queued, resulting in
The defining example of a circuit-switched network variable delay and throughput depending on the traffic
is the early analog telephone network. When a call is load in the network.
made from one telephone to another, switches within Packet switching contrasts with another principal
the telephone exchanges create a continuous wire circuit networking paradigm, circuit switching, a method which
between the two telephones, for as long as the call lasts. sets up a limited number of dedicated connections of
Circuit switching contrasts with packet switching constant bit rate and constant delay between nodes for
which divides the data to be transmitted into packets exclusive use during the communication session. In case
transmitted through the network independently. Packet of traffic fees (as opposed to flat rate), for example, in
switching shares available network bandwidth between cellular communication services, circuit switching is
multiple communication sessions. characterized by a fee per time unit of connection time,
In circuit switching, the bit delay is constant during even when no data is transferred, while packet switching
a connection, as opposed to packet switching, where is characterized by a fee per unit of information.
102 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Two major packet switching modes exist: network is assigned a length based on whatever qualities
(1) Connectionless packet switching, also known are considered important to that network. If the link
as datagram switching, and between two routers is half duplex or full-duplex, the
(2) Connection-oriented packet switching, also length of the link in one direction might be different
known as virtual circuit switching. from the length of the link in the other directions. The
In the first case, each packet includes complete physical distance that the signal has to travel is not
addressing or routing information. The packets are changed but other factors such as traffic load or quality
routed individually, sometimes resulting in different of the cable may differ.
paths and out-of-order delivery. In the second case, a Non-Adaptive Routing: In some routing protocols
connection is defined and preallocated in each involved once a pathway to a destination has been selected, the
node during a connection phase before any packet is router sends all packets to that destination along that
transferred. The packets include a connection identifier one route. In other words, the routing decisions are not
rather than address information, and are delivered in made based on the condition or topology of the
order. See below. networks.
Packet mode communication may be utilized with Adaptive Routing: Other routing protocols employ
or without intermediate forwarding nodes (packet a technique called adaptive routing, by which a router
switches or routers). In all packet mode communication, may select a new route for each packet in response to
network resources are managed by statistical changes in condition and topology of the networks.
multiplexing or dynamic bandwidth allocation in which Given a transmission from network A to network D. A
a communication channel is effectively divided into an router may send the first packet by way of network B,
arbitrary number of logical variable-bit-rate channels the second packet by way of network C and the third
or data streams. Statistical multiplexing, packet packet by way of network Q depending on which route
switching and other store-and-forward buffering is most efficient at the moment.
introduces varying latency and throughput in the CONGESTION CONTROL
transmission. Each logical stream consists of a sequence When too many packets are present in the subnet,
of packets, which normally are forwarded by the performance, degrades. This situation is called
multiplexers and intermediate network nodes congestion. In figure depicts the symptom when the
asynchronously using first-in, first-out buffering. number of packets dumped into the subnet by the hasts
Alternatively, the packets may be forwarded according is within its carrying capacity, they are all delivered and
to some scheduling discipline for fair queuing, traffic no delivered is proportional to the number sent.
shaping or for differentiated or guaranteed quality of However, as traffic increases too far, the routers are no
service, such as weighted fair queuing or leaky bucket. longer able to cape and they begin losing packets. This
In case of a shared physical medium, the packets may tends to make matter worse. At very high traffic
be delivered according to some packet-mode multiple performance callapses completely and almost no packets
access scheme. are delivered.
ROUTING Perfect
In routing, the term shortest can mean the Maximum Carrying
combination of many factors including shortest, Capacity of Subject Desirable
cheapest, fastest, most reliable and so on. As we have
seen, the job of routers is to forward packets through a
Packet delivered

set of networks. For example, that we want to move a

packet from network A to network C via router B. Often
howear, more than one pathway exists between the point
of origin and the point of destination. For example, the Congested
packet could reach network C by going through router
D instead of router B, or possibly even going directly
from A to C. Whenever, there are multiple options, the Packets Sent
router chooses the pathway.
We can combine all of the factors affecting a link In contrast, closed loop solutions are based on the
into one number and call that number the symbolic concept of a feedback loop. This approach has three
length of the link. In some protocols, each link in a parts when applied to congestion control.
 NETWORK LAYER / 103

1. Monitor the system to detect when and where

congestion occurs. B
2. Pass the information to places where action can Net:14 Net:50
be taken.
3. Adjust system operation to correct the problem. A C
ROUTING PROTOCOLS: DISTANCE VECTOR
Net:68 F Net:62
VS LINK-STATE ROUTING
Routing Protocols: In routing the pathway with Net:25 Net:72
the lowest cast is considered the best. As long as the
cast of each link is known a router can find the optimal E Net:10 D
combination for any transmission. Several routing
algerithm exists for making these calculations. The most
Link State Routing: The keys to undrstanding link
popular are distance vector routing and link state routing.
state routing are different from those in distance vector
Distance Vector Routing: In distance vector routing. In link state routing each router shares its
routing, each router periodically shares its knowledge knowledge of its neighbourhood with every other routers
about the entire network with its neighbours. The three in the internetwork. The following are true of link state
key to understanding how this algorithm works are as routing:
follows: 1. Knowledge about the Neighbourhood: Instead
1. Knowledge about the Whole Network: Each of sending its entire routing table a router sends
router shares its knowledge about the entire network. It information about its neighbourhood only.
sends all of its collected knowledge about the network 2. To All Routers: Each router sends this
to its neighbours. At the outset, a router knowledge of information to every other router on the internetwork,
the nework may be sparse. How much it knows however, not just to its neighbours. It does so by a process called
is unimportant, it sends whatever it has. flooding. Hooding means that a router sends its
2. Routing only to Neighbours: Each router information to all of its neighbours. Each neighbours
periodically sends its knowledge about the network only sends the packet to all of its neighbours and so on. Every
router that receives the packet sends copies all of its
to those routers to which it has direct links. It sends
neighbours. Finally, evey router receives a copy of the
whatever knowledge its has about the whole network
same information.
through all of its ports. This information is received and
3. Information Sharing when there is a Change:
kept by each neighbouring router and used to update
Each router sends out information about the neighbours
that router own information about the network. when there is a change.
3. Information Sharing at Regular Intervals: For
example, every 30 seconds, each router sends its
4 B 2
information about the whole network to its neighbours.
Net:04 Net:45
This sharing occurs whether or not the network has 1 5

changed since the last time information was exchanged. A C

Distance vector routing simplifies the routing 2 3
Net:68 F Net:82
process by assuming a cast of one unit for every link. In 2 2
this way, the efficiency of transimssion is a function Net:13 Net:55
only of the number of links required to reach a
destination. In distance vector routing, the cast is based 3 E 2
Net:08
3
D 5
on hap count.
104 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

DV PROBLEMS router loses a link, the router advertises the loss of a

Convergence: The term convergence refers to the route to its neighbour device. Route poisoning enables
time it takes for all of the routers to understand the the receiving router to advertise a route back toward
current topology of the network. When a router receives the source with a metric higher than the maximum. The
an update from a neighbouring router, it compares the advertisement back seems to violate split horizon, but
update to its own routing table.The router adds the cost it lets the router know that the update about the down
of reaching the neighbouring router to the path cost network was received. The router that received the
reported by the neighbour to establish the new metric. update also sets a table entry that keeps the network
If the router learns about a better route (smaller total state consistent while other routers gradually converge
metric) to a network from its neighbour, the correctly on the topology change. This mechanism
routerupdates its own routing table. It’s too time allows the router to learn quickly of the down route and
consuming process. Because in a 10 router topology to ignore other updates that might be wrong for the hold-
last router will know about the network of first router down period. This prevents routing loops.
only while all middle router will complete their periodic A poisoned route has an infinite metric assigned to
update. For example, if interval timer is set to 60 second it. A poison reverse causes the router to break split
then last router will know about first network in 60*8 horizon rule and advertise the poisoned route out all
480 second or 8 minute. interfaces. When a router detects that one of its
A Solution is to Implement Triggered Updates: connected routes has failed, the router will poison the
The distance vector routing protocol would still generate route by assigning an infinite metric to it. In IP RIP, the
periodic updates; however, whenever a change takes route is assigned a hop count of 16 (15 is the maximum),
place, the router will immediately generate an update thus making it an unreachable network. When a router
without waiting for the periodic timer to expire. This advertises a poised route to its neighbors, its neighbours
can decrease convergence times, but it also creates a break the rule of split horizon and send back to the
problem. If you have a flapping route, then an update originator the same poisoned route, called a poison
will be triggered each time the route changes state, which reverse. This ensures that everyone received the original
creates a lot of unnecessary broadcast traffic in your
update of the poisoned route.
network and could cause a broadcast storm.
NETWORK ADDRESSING
Routing Loops: A routing loop is a layer-3 loop in
The physical addresses that identify individual
the network. Basically, it is a disagreement about how
devices, the internet requires an additional addressing
to reach a destination network. Because distance vector
convention: an address that identifies the connection of
routing protocols trust the next router without compiling
a hast to its network.
a topology map of all networks and routers, distance
Each internet address consists of four bytes (32 bits)
vector protocols run the risk of creating loops in a
defining three fields: class id, net id and hast id. These
network. This is analogous of driving to a location
parts are of varying lenghts depending on the class of
without a map. Instead, you trust what each sign tells
the address.
you. Trusting the street signs might get you where you
want to go, but I’ve been in some cities where trusting
what the signs say will lead you in loops. The same is
true with distance vector routing protocols. Simply
trusting what the next router tells it can potentially lead
the packets to loop endlessly. These loops could saturate
a network and cause systems to crash. This, in turn, An Internet address is made of four bytes that define
makes managers very upset and means that you have to a hast connection to a network.
work late into the evening to fix it. There are currently five different field length
Solution: Route Poisoning: Another operation patterns in use, each defining a class of address. Class
complementary to split horizon is a technique called A address are numerically the lowest. They use only
route poisoning. Route poisoning attempts to improve one byte to identify class type and netid and leave three
convergence time and eliminate routing loops caused bytes available for hastid numbers. Class B and Class
by inconsistent updates. With this technique, when a C networks which provides two and one byte hastid
 NETWORK LAYER / 105

fields respectively. Currently both class A and class B Every IP packet has an IP (Internet Protocol) header
are full. Addresses are available in class C only. Class that stores information about the packet, including:
D is reserved for multicast addresses. Version
IHL
bytes bytes 2 bytes 3 bytes 4 Type of Service
0 Netid Hostid Total Length
Identification
10 Netid Flags
Fragment Offset
110 Netid Hostid
Time to Live
Protocol
1110 Multicast address
Header Checksum
Source Address
1111 Reserved for future use
Destination Address
Internet classes Options
Three of these fields are involved in packet
fragmentation.
Identification
Flags
Fragment Offset
Identification: 16 bits: An identifying value that
the sender assigns to aid in assembling a datagram’s
fragments.
Flags: 3 bits: Various Control Flags.
Bit 0: reserved, must be zero
Bit 1: (DF) 0 = May Fragment, 1 = Don’t Fragment.
FORWARDING Bit 2: (MF) 0 = Last Fragment, 1 = More
It is sometimes useful to make a distinction between Fragments.
routing which is making the decision which routes to Fragment Offset: 13 bits: This field indicates
use and forwarding, which is what happens when a where in the datagram this fragment belongs.
packet arrives. One can think of a router as having two The fragment offset is measured in units of 8 octets
process inside it. One of them handles each packet as it (64 bits). The first fragment has offset zero.
arrives, looking up the outgoing line to use for it in the Much like the IP header, the TCP (Transmission
routing tables. This process is forwarding. The other Control Protocol) header stores information about the
process is responsible for filling in and updating the packet:
routing tables. Source Port
Regardless of whether routes are chosen Destination Port
independently for each packet or only when new Sequence Number
connection are established certain properties are Acknowledgement Number
desirable in a routing algorithm: correctness, simplicity, Data Offset
robustness, stability and optimality. Flags
FRAGMENTATION Window
Every packet based network has an MTU Checksum
(Maximum Transmission Unit) size. The MTU is the Urgent Pointer
size of the largest packet that that network can transmit. Options
Packets larger than the allowable MTU must be Padding
divided into smaller packets or fragments to enable them Packet fragmentation can be utilized to get around
to traverse the network. blocking rules on some firewalls.
106 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

This is done by cheating with the value of the probably function normally, although this error message
Fragment Offset. The trick is to set the Fragment Offset’s will become an annoyance as it pops up repeatedly until
value on the second packet so low that instead of the issue is addressed.
appending the second packet to the first packet, it 2. IP Address Conflict (or Address Already in
actually overwrites the data and part of the TCP header Use): If a computer is set up with a static IP address
of the first packet. that is being used by some other device on the network,
If someone wants to ‘telnet’ into a network where a the computer (and possibly also the other device) will
packet filtering firewall blocks TCP port 23, SMTP port be unable to use the network. This problem can even
25 is allowed into that network. occur with dynamic (DHCP) addressing in some cases.
The user would have to send two packets: 3. The Network Path cannot be Found: Updating
the TCP/IP configuration can resolve this issue
The first packet would:
sometimes seen on Microsoft Windows computers when
Have a Fragmentation Offset of 0.
trying to access another device via Network Neighbour-
Have a DF bit equal to 0 to mean “May
hood.
Fragment” and an MF bit equal to 1 to mean 4. Duplicate Name Exists on the Network: After
“More Fragments.” starting up a Microsoft Windows computer connected
Have a Destination Port in the TCP header of to a local network, you may encounter this error as a
25. TCP port 25 is allowed, so the firewall would balloon message. When it occurs, your computer will
allow that packet to enter the network. be unable to access the network. You may need to change
The second packet would: the name of your computer to resolve this issue.
Have a Fragmentation Offset of 1. This means 5. Limited or No Connectivity: When attempting
that the second packet would actually overwrite to access a Web page or network resource on a Windows
everything but the first 8 bits of the first packet. shared drive, you may receive a pop-up dialog error
Have a DF bit equal to 0 to mean “May message that starts with the words “limited or no
Fragment” and an MF bit equal to 0 to mean connectivity.”
“Last Fragment.” 6. Connected with Limited Access: A technical
Have a Destination Port in the TCP header of glitch in Windows Vista causes this error message to
23. This would normally be blocked, but will appear when making certain types of wireless
not be in this case! connections. Microsoft has provided a fix for this issue
The packet filtering firewall will see that the in Vista service packs. This error can also occur on a
Fragment Offset is greater than zero on the second home network for other reasons that may require you to
packet. From this data, it will deduce that the second reset your router.
packet is a fragment of another packet and it will not 7. Unable to Join Network Failure (error-3): This
check the second packet against the rule set. error appears on the Apple iPhone (or iPod Touch) when
When the two packets arrive at the target host, they it fails to join a Wi-Fi network. You can troubleshoot
will be reassembled. The second packet will overwrite this problem the same as you would for a PC that cannot
most of the first packet and the contents of the combined connect to a hotspot.
packet will go to port 23. 8. Unable to Establish the VPN Connection
(error 800): When using a VPN client on Windows
ERROR MESSAGING SERVICES
XP, you may receive error 800 when trying to connect
If your network connection is not configured
to the VPN server. This generic message can indicate
properly or suffers a technical failure, you will often
problems on either the client or server side.
see some error message generated and displayed on your
screen. These messages give helpful clues to the nature
of the issue. Use this list of the common network-related Q. 1. What is custom subneting?
error messages to help troubleshoot and fix networking Ans. Subnets that can be customized; i.e. modifying
problems. the dividing point between subnet ID and host ID to
1. A Network Cable is Unplugged: This message suit the needs of our network. The subnet mask that we
appears as a Windows desktop balloon. If your use when creating a customized subnet is, called a
connection is wired, you may lose access to the network. custom subnet mask. This custom subnet mask is used
If your connection is wireless, your network will to find the customization.
 NETWORK LAYER / 107

Q. 2. What is Data Fragmentation? Explain how Leftmost 3 bits: 110

Data Fragmentation works? The next 21 bits define network.
Ans. Fragmentation occurs when storage space is 8 bits define the hostid.
used inefficiently due to which storage capacity and Class D:
performance is reduced. First 4 bits: 1110
Data fragmentation occurs when a large object is The remaining 28 bits define multicast addresses.
inserted into storage that has already suffered external No netid or hostid in a class D address.
fragmentation due to which the data object is broken up Addresses is used for multicasting.
into many pieces that are not close together. Class E:
When free storage becomes divided into many First 4 bits: 1111
small pieces over time, its called External fragmentation. Addresses reserved for special use on the Internet.
Q. 3. What is Piconet? There is no netid or hostid in a class E address.
Ans. Piconet is an ad-hoc network by linking a Q. 6. What is Subneting? Explain the advantages
group of users which uses blue tooth technology of using Subneting.
protocols for allowing one ‘master’ device to Ans. Subneting is dividing a network into several
interconnect with up to seven active ‘slave’ devices. subnets.
Further, up to 255 slave devices could be inactive or This is usually done for the following purposes:
packed and the master device can bring into active status Reducing network traffic by decreasing the
at any given point of time. number of broadcasts.
The range of Piconet will vary based on the blue Exceeding the limitations in a local area
tooth device class. The data transfer rate varies around network.
200 to 2100 KB/S, depending upon the synchronous or Enabling people to connect to the network.
asynchronous connection usage, and the number of remotely without opening the entire network.
devices connected in the piconet. Q. 7. Explain the functionality of PING.
Q. 4. What is meant by the term ‘IP Spoofing’? Ans. Ping is particularly used to check if the system
Ans. IP Spoofing: Sending messages by intruder is in network or not. It also gives packet lost information.
to a computer system with an IP address that indicates In windows ping command is written as ping ip_address.
the message is sent from a trusted host, for gaining The output returns the data packets information. The
unauthorized access to computer systems. The hacker / number of packets sent, received and lost is returned by
intruder first utilize a variety of techniques for finding PING.
an IP address of a trusted host. Later he modifies the Q. 8. Explain the core naming mechanism,
packet headers, in order to give an illusion that the Domain Name System (DNS).
packets are sent from the host. Ans. A Domain Name system is used to convert
Q. 5. Explain the classes of IP addresses. Why the names of the website on the internet to IP addresses.
do we need them? The domain names for each IP addresses are stored in a
Ans. Class A: database that is distributed across different servers. A
Range: from 0.0.0.0 to 127.255.255.255. domain name space consists of a tree of domain names.
Leftmost bit: 0. The tree has zones. Zones consist of a collection of
First 8 bits: netid. connected nodes. These nodes are served by a name
Remaining 24 bits: hostid. server. A domain name is usually in the form of
Class B: mydomain.com. Here, .com is the top level domain.
Range: from 128.0.0.0 to 191.255.255.255. Where as mydomain is the sub domain or subdivision.
Leftmost 2 bits: 10 A host name is a domain name that has one or more IP
First 16 bits: netid addresses associated with it.
Last 16 bits: the hostid Q. 9. Explain the importance of increasing and
Class C: decreasing subnet bits.
Range: from 192.0.0.0 to 223.255.255.255. Ans. Adding 1’s as a bit means increasing the
Class C networks use the first 24 bits to determine subnets and decreasing the hosts. Removing or
the netid. decreasing ’s means decreasing subnets and increasing
108 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

hosts. Hence, by increasing or decreasing the subnet device to have a unique IP address for each network
architecture can be decided as per needs. connection (network interface). Every IP packet contains
Q. 10. Why do we need a subnet mask? a source IP address and a destination IP address. As a
Ans. A subnet mask allows identification of host device moves from one network to another, its IP address
part and network part of an IP address. Subnet mask changes.
can be used to find if an IP address is present on a subnet Computers using the TCP/IP for communication
or not. are uniquely identified by a 32 bit address called as an
Q. 11. What is a MAC Address? IP address. The routers use the IP address information
Ans. MAC (Media Access Control) addresses are to forward the packet to the destination computer.
globally unique addressed that are written into hardware IP addresses are categorized as:
at the time of manufacture. The MAC address is a unique Private Address: These IP addresses are used
value associated with a network adapter. MAC addresses exclusively within a private network and not for public
are also known as hardware addresses or physical to see.
addresses. They uniquely identify an adapter on a LAN. Public Address: These are registered IP addresses
MAC addresses are 12-digit hexadecimal numbers (48 used for public.
bits in length). Each IP address has a network address and a host
Q. 12. MAC vs. IP Addressing. address. IP addresses are expressed in four sets of three
Ans. Whereas MAC addressing works at the data numbers, separated with dots. Each set is called as an
link layer, IP addressing functions at the network layer octet because when converted to binary; it denotes eight
(layer 3). It’s a slight oversimplification, but one can binary.
think of IP addressing as supporting the software Q. 16. What is the difference between Static and
implementation and MAC addresses as supporting the Dynamic IP.
hardware implementation of the network stack. The Ans. Static IP is also called as permanent address
MAC address generally remains fixed and follows the assigned to each device in a network, whereas Dynamic
network device, but the IP address changes as the IP, a temporary address assigned to the device via DHCP
network device moves from one network to another. software. IP address assigned to your service by your
Q. 13. Define Spanning-Tree Protocol (STP). cable or DSL Internet provider is typically dynamic IP.
Ans. Spanning-Tree Protocol (STP) as defined in In routers and operating systems, the default
the IEEE 802.1D is a link management protocol that configuration for clients is dynamic IP.
provides path redundancy while preventing undesirable Q. 17. What is the difference between public and
loops in the network. For an Ethernet network to function private IP?
properly, only one active path can exist between two Ans. A public IP address allows equipment
stations. Loops occur in networks for a variety of accessible to everyone on the internet. A private IP
reasons. The most common reason you find loops in address is for private use within the network and allows
networks is the result of a deliberate attempt to provide many more PCs to be connected. If you are using a
redundancy – in case one link or switch fails, another
private IP and wants VOIP, you need to change to a
link or switch can take over.
public IP address.
Q. 14. What is VPN?
Q. 18. What is Network Address Translation?
Ans. A VPN is a service that offers secure, reliable
Ans. Network Address Translation acts as an agent
connectivity over a shared public network infrastructure
between the Internet and a local network. It is a dynamic
such as the Internet. VPNs maintain the same security
and management policies as a private network. They method which is used to minimize Internet connectivity
are the most cost effective method of establishing a needs. Network address translation describes the
virtual point-to-point connection between remote users rewriting of the Internet Protocol (IP) addresses of data
and an enterprise customer’s network. packets so that multiple transmissions require only one
Q. 15. How would you define IP address? IP address.
Ans. IP address or Internet Protocol address is the Q. 19. Define IP multicast.
address of a device attached to an IP network (TCP/IP Ans. IP multicast technology reduces traffic by
network). It is a must for every client, server and network sending stream of information to many recipients at one
 NETWORK LAYER / 109

go. Video conferencing, stock quotas are the examples where these packets will travel. When a packet is
based on IP multicast. received, a network device examines the packet and
Q. 20. What is subneting? matches it to the routing table entry providing the best
Ans. Subnet adds one level to the way IP address match for its destination. The packet keeps hopping until
is represented. It logically organizes the network. For it reaches its destination.
instance, it can logically group computers belongs to Q. 25. Explain the structure and use of internet
the finance department. addresses.
Q. 21. Define Address Resolution Protocol. Ans. Each IP address is 32 bit long. In human
Ans. Address Resolution Protocol ARP, is language the IP addresses are written in dotted decimal
responsible for mapping an IP address to its notation. These are then converted to binary by the
corresponding physical network address. It is mostly computer. Each IP address has two parts: Network
seen on Ethernet network. identifier or a network ID and host ID. The current
Q. 22. What is Routing table? internet protocol standard is IPV4. The IP addresses
Ans. A routing table stores the routes of the various are divided into three classes: a class A network, a class
nodes in a network. Nodes can be any electronic device B network, and a class C network. Class A being the
connected to the network. The table is usually stored in largest. The four digit numbers in an IPV4 address, each
network of class A will have different first number, and
a router or the network computer as a database or file.
then its network will be addressed by the rest of the
This information helps to fond the best possible path.
three numbers, or three bytes. The IP addresses identify
The routing table has at least 3 fields: the destination
a machine to deliver packets and load web pages.
network id, cost of the path, next hop or address to send
Q. 26. Explain how names are translated
the packet.
(resolved) into IP address.
Q. 23. What is Distance Vector Routing
Ans. Domain Name server or DNS is used to
Protocols?
resolve names into IP addresses. When a web address
Ans. The main goal of Distance Vector Routing is entered into the browser, the DNS client sends a
Protocols is to find out the best path for, the data packet request to the DNS server to find the corresponding IP
to reach the destination. Distance here could be the hops. address for the name. The DNS server receives this
The three different types of Distance Vector routing request and searches for the corresponding IP address
protocols include: Routing Information Protocol (RIP in the database. If at this point the resolution fails, this
v1 and v2) and Interior Gateway Routing Protocol. The server sends this request to the parent server. The request
protocol is easy to manage however, not well scalable. keeps going up the hierarchy to the parent servers or
The Distance Vector protocol initially prepares a the closest authoritative of the DNS server to resolve
Routing table which is shared with other routers. This the address. If the request times out an error is retuned
routing table is shared between routers present in the to the client. If the server is able to resolve the name
same network. A new routing table is prepared when requested, it passes the information back to the client.
some new information is received from some other The next request sent by the client is to request for a
router. Now, the bad routing paths are removed keeping web page for the IP address.
only the smallest hop paths. This new table is then Q. 27. Describe the basics of internet routing.
communicated to other routers. Ans. When a source sends a packet to a destination,
Q. 24. What is Routing Protocol? this packet has a specific path or route it follows.
Ans. Routing protocol is the way to send routing Different routing protocols are used to find the shortest
information between any routers in an autonomous path to the destination. The protocols maintain routing
system. tables. Routing tables consist of a set of rules used to
When a source sends a packet to a destination, this determine where these packets will travel. When a
packet has a specific path or route it follows. Different packet is received, a network device examines the packet
routing protocols are used to find the shortest path to and matches it to the routing table entry providing the
the destination. The protocols maintain routing tables. best match for its destination. The packet keeps hopping
Routing tables consist of a set of rules used to determine until it reaches its destination.
110

Transport Layer
10

The Transport layer is regarded as the highest of ADDRESSING AND MULTIPLEXING

the lower layer protocols in the OSI protocol stack. It Addressing, as is expected, is there so that the user
resides above the Physical, Data-Link and Network of one transport entity can establish a connection with
layers and just below the session layer and is concerned or make a connectionless data transfer to a user of
with the transmission of data between end systems across another transport entity. To do this, the destination
a communications facility. It provides a messaging transport entity must have an address. The address is
service for the Session layer and hides the underlying made up of the following:
network from the upper layers. As such it may need to User identification
be very complex in order to deal with a variety of Transport entity identification
network characteristics and capabilities. End-system identification
The Transport layer provides the means to establish, A TSAP (Transport Service Access Point)
maintain, and release transport connections on behalf represents the user identification. Each
of session entities. It provides reliable end-to-end data individual user of a transport layer accesses
transport. Error checking and other reliability features the services of the transport layer via a unique
are handled by the protocols in the Transport layer if TSAP. A destination receipt of a TSAP;
they are not handled by the underlying networks. This communicates it to the transport entity so that
is the case, for example, with Frame Relay and
received data can be routed to the proper
Asynchronous Transfer Mode (ATM) networks, which
transport user.
do not perform error checks when delivering packets.
The transport entity identification is rarely
Instead, error checking is assumed to be handled by
needed due to there being only a single
Transport layer protocols which examine information
transport entity in each end system. A specific
in the packet to make sure the information was delivered
identification is not necessary here.
error-free.
An NSAP (Network Service Access Point) is
The Transport layer can provide a connection-
oriented or connectionless service. In a connection- the end-system identification. This
oriented session, a circuit is established through which identification is passed down to the network
packets flow to the destination. In this arrangement, service where it is used in the end-system
packets arrive in order and do not require a full address identification.
or other information because the circuit guarantees their The user must know the address it requires ahead
delivery to the proper destination. Connectionless of time. This is simply a system configuration function.
session does not establish circuits or provide reliable A Specific Case: Certain processes may be only
data delivery. Packets are fully addressed and sent out applicable to a certain number of transport users. For
over the network. The Transport layer protocols at the example, a process which collects statistics on
destination can re-order the packets which arrive out of performance in a station. A central network management
order and request retransmission of missing or defective routine may wish to connect to the process to obtain
packets. statistics. These processes should be well known only
 TRANSPORT LAYER / 111

to that certain number of users to which they are Flow control is a good example of a protocol
applicable. function that must be implemented in several layers of
Well-known Addresses are given to Common the OSI architecture model. At the transport level flow
Services control will allow the transport protocol entity in a host
A Specific Case: Services such as word processing to restrict the flow of data over a logical connection
and time sharing would have addresses known to all. from the transport protocol entity in another host.
A name server exists which, when a user requests a However, one of the services of the network level is to
service by a generic or global name; looks up its prevent congestion. Thus the network level also uses
directory and returns an address. flow control to restrict the flow of network protocol
A Specific Case: Commonly used services may data units (NPDUs).
change their address occasionally. For example, a data The flow control mechanisms used in the transport
entry process may be moved from one station to another layer vary for the different classes of service. Since the
on a local network to balance load. The names of these different classes of service are determined by the quality
moving processes can be kept in the name server which of service of the underlying data network which
updates addresses when they change. transports the transport protocol data units (TPDUs), it
In certain cases, the destination user is to be a is these which influence the type of flow control used.
process which is spawned at request time. The initiating Thus flow control becomes a much more complex
user can send a process request to a well known address issue at the transport layer than at lower levels like the
The user at that address is a privileged system process datalink level.
Two reasons for this are:
that will spawn the new process and return an address.
Flow control must interact with transport users,
A Specific Case: This might be applied where a
transport entities, and the network service.
user has a private application that will execute on a
Long and variable transmission delays between
remote mainframe but be invoked from a local
transport entities.
minicomputer. For example, an RJE-type request can
Flow control causes queuing amongst transport
be issued to a remote job-management process that
users, entities, and the network service. We take a look
spawns the simulator process.
at the four possible queues that form and what control
Multiplexing: Multiplexing have several conver-
policies are at work here.
sation into connections, virtual circuits and physical
The transport entity is responsible for generating
links plays a role in several layers of the network
one or more transport protocol data units (TPDUs) for
architecture. In transport layers, the need for passing onto the network layer. The network layer
multiplexing can aries in a no. of ways. For example, if delivers the TPDUs to the receiving transport entity
only one network address is available on a hast, all which then takes out the data and passes it on to the
transport connections on that machine have to use it, destination user. There are two reasons why the receiving
when a TPBV comes in come way is needed to tell which transport entity would want to control the flow of
process to give it to. This situation called up word TPDUs:
multiplexing. The receiving user cannot keep up with the
Multiplexing can also be useful in the transport flow of data.
layer for another reason. For example, that a subnet uses The receiving transport entity itself cannot
virtual circuits internally and imposes a maximum data keep up with the flow of TPDUs.
rate on each one. If a user needs more bandwidth then When we say that a user or transport entity cannot
one virtual circuits traffic among them on a round robin keep up with the data flow, we mean that the receiving
basis. This modus operand is called downword buffers are filling too quickly and will overflow and
multiplexing. lose data unless the rate of incoming data is slowed.
FLOW CONTROL Four possible ways to cope with the problem are:
Flow control is a function for the control of the Let it be and do nothing
data flow within an OSI layer or between adjacent layers. Refuse any more TPDUs from the network
In other words it limits the amount of data transmitted service
by the sending transport entity to a level, or rate, that Use a fixed sliding-window protocol
the receiver can manage. Use a credit scheme
112 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

There are different issues to be considered with allows the use of different, but generally TCP-friendly
transport flow control over different levels of network congestion control mechanisms.
service. The more unreliable the network service A DCCP connection contains acknowledgement
provided the more complex flow control mechanism traffic as well as data traffic. Acknowledgments inform
that may be needed to be used by the Transport layer. a sender whether its packets have arrived, and whether
they were marked by Explicit Congestion Notification
The credit scheme works well with the different network
(ECN). Acks are transmitted as reliably as the congestion
services although specific issues need to be addressed control mechanism in use requires, possibly completely
as with a Reliable Nonsequencing Network Service and reliably.
an Unreliable Network Service. DCCP has the option for very long (48-bit)
The credit scheme seems most suited for flow sequence numbers corresponding to a packet ID, rather
control in the transport layer with all types of network than a byte ID as in TCP.
service. It gives the receiver the best control over data DATA TRANSPORT
flow and helps provide a smooth traffic flow. Sequence The transport layer is responsible for delivering data
to the appropriate application process on the host
numbering of credit allocations handles the arrival of
computers.The protocol layer just above the Internet
ACK/CREDIT TPDUs out of order, and a window timer Layer is the Host-to-Host Transport Layer. The two most
will ensure deadlock does not occur in a network important protocols in the Transport Layer are
environment where TPDUs can be lost. Transmission Control Protocol (TCP) and User
CONGESTION CONTROL Datagram Protocol (UDP). TCP provides reliable data
The Datagram Congestion Control Protocol delivery service with end-to-end error detection and
(DCCP) is a message-oriented Transport Layer protocol. correction. UDP provides low-overhead, connectionless
DCCP implements reliable connection set up, teardown, datagram delivery service. Both protocols deliver data
between the Application Layer and the Internet Layer.
Explicit Congestion Notification (ECN), congestion Applications programmers can choose whichever
control, and feature negotiation. DCCP was published service is more appropriate for their specific
as RFC 4340, a proposed standard, by the IETF in applications.
March, 2006. User Datagram Protocol: The User Datagram
DCCP provides a way to gain access to congestion Protocol gives application programmes direct access to
control mechanisms without having to implement them a datagram delivery service, like the delivery service
at the Application Layer. It allows for flow-based that IP provides. This allows applications to exchange
messages over the network with a minimum of protocol
semantics like in Transmission Control Protocol (TCP),
overhead.
but does not provide reliable in-order delivery. UDP is an unreliable, connectionless datagram
Sequenced delivery within multiple streams as in the protocol. As noted previously, “unreliable” merely
Stream Control Transmission Protocol (SCTP) is not means that there are no techniques in the protocol for
available in DCCP. verifying that the data reached the other end of the
DCCP is useful for applications with timing network correctly. Within your computer, UDP will
constraints on the delivery of data. Such applications deliver data correctly. UDP uses 16-bit Source Port and
Destination Port numbers in word 1 of the message
include streaming media, Multiplayer online games and
header, to deliver data to the correct applications
Internet telephony. Primary feature of these applications process.
is that old messages quickly become stale so that getting UDP message format:
new messages is preferred to resending lost messages.
Currently such applications have often either settled for Bits
TCP or used User Datagram Protocol (UDP) and
implemented their own congestion control mechanisms,
or have no congestion control at all.
While being useful for these applications, DCCP
can also be positioned as a general congestion control
mechanism for UDP-based applications, by adding, as
needed, a mechanism for reliable and/or in-order
delivery on the top of UDP/DCCP. In this context, DCCP
 TRANSPORT LAYER / 113

Why do applications programmers choose UDP as Must be able to adapt to congestion in the
a data transport service? There are a number of good application layer
reasons. If the amount of data being transmitted is small, Transmission Control Protocol: Applications that
the overhead of creating connections and ensuring require the transport protocol to provide reliable data
reliable delivery may be greater than the work of re- delivery use TCP because it verifies that data is delivered
transmitting the entire data set. In this case, UDP is the across the network accurately and in the proper
most efficient choice for a Transport Layer protocol. sequence. TCP is a reliable, connection-oriented, byte-
Applications that fit a query-response model are also stream protocol. Let’s look at each of the terms–reliable,
excellent candidates for using UDP. The response can connection-oriented, and byte-stream–in more detail.
be used as a positive acknowledgment to the query. If a TCP provides reliability with a mechanism called
response isn’t received within a certain time period, the Positive Acknowledgement with Re-transmission
application just sends another query. Still other (PAR). Simply stated, a system using PAR sends the
applications provide their own techniques for reliable data again, unless it hears from the remote system that
data delivery, and don’t require that service from the the data arrived okay. The unit of data exchanged
transport layer protocol. Imposing another layer of between cooperating TCP modules is called a segment.
acknowledgment on any of these types of applications Each segment contains a checksum that the recipient
is inefficient. uses to verify that the data is undamaged. If the data
UDP Applications: segment is received undamaged, the receiver sends a
Useful for applications that prefer timeliness positive acknowledgement back to the sender. If the data
to reliability segment is damaged, the receiver discards it. After an
Voice-over-IP appropriate time-out period, the sending TCP module
Streaming video re-transmits any segment for which no positive
Must be able to tolerate some loss of data acknowledgment has been received.
TCP segment format

Bits

TCP is connection-oriented. It establishes a logical end-to-end connection between the two communicating
hosts.
TCP Applications:
Useful for applications that require reliable data delivery, and can tolerate some timing variation
File transfer and web downloads
Email
Instant messaging
114 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

PORT NUMBERS After IP passes incoming data to the transport

A host may have many TCP and UDP connections protocol, the transport protocol passes data to the correct
at any time. Connections to a host are distinguished by application process. Application processes are identified
a port number, which serves as a sort of mailbox number by port numbers, which are 16-bit values. The source
for incoming datagrams. There may be many processes port number, which identifies the process that sent the
using TCP and UDP on a single machine, and the port data, and the destination port number, which identifies
numbers distinguish these processes for incoming the process that is to receive the data are contained in
packets. When a user program opens a TCP or UDP the header of each TCP segment and UDP packet.
socket, it gets connected to a port on the local host. The Port numbers are not unique between transport layer
application may specify the port, usually when trying protocols, the numbers are only unique within a specific
to reach some service with a well-defined port number, transport protocol. It is the combination of protocol and
or it may allow the operating system to fill in the port port numbers that uniquely identifies the specific process
number with the next available free port number. the data should be delivered to.

Figure shows data packets multiplexed via TCP or In figure, the source host randomly generates a
UDP through port addresses and onto the targeted TCP/ source port, in this example 3044. It sends out a segment
IP applications. with a source port of 3044 and a destination port of 23.
In figure, if a data packet arrives specifying a The destination host receives the segment, and responds
transport protocol of 6, it is forwarded to the TCP back using 23 as it source port and 3044 as its destination
implementation. If the packet specifies 17 as the required port.
protocol, the IP layer would forward the packet to the SERVICE MODELS
programs implementing UDP. DCCP: Datagram Congestion Control Protocol
The Datagram Congestion Control Protocol
(DCCP) is a message-oriented transport layer protocol.
DCCP provides bidirectional unicast connections of
congestion-controlled unreliable datagrams. DCCP
implements reliable connection set up, teardown, ECN,
congestion control, and feature negotiation. DCCP is
suitable for applications that transfer fairly large
amounts of data and that can benefit from control over
the tradeoff between timeliness and reliability.
Stream Control Transmission Protocol: SCTP is
a reliable transport protocol operating over IP. SCTP is
more akin to TCP than UDP, however, it yields
Figure shows the exchange of port numbers during additional features to TCP while still supporting much
the TCP handshake. of the same functionality. So SCTP is connection-
 TRANSPORT LAYER / 115

oriented and implements the same congestion/flow between a user and network. The following are some
control. Detection of data corruption, loss of data and user related attributes:
duplication of data is achieved by using checksums and SCR: The sustained cell rate (SCR) is the average
sequence numbers. A selective retransmission cell rate over a long time interval.
mechanism is applied to correct loss or corruption of PCR: The Peak cell rate (PCR) defines the sender’s
data. maximum cell rate.
INTRO TO RELIABILITY MCR: The Minimum cell rate (MCR) defines the
Reliability specify the guarantees that the protocol minimum cell rate acceptable to the sender.
provides with respect to the delivery of messages to the CVDT: The Cell variation delay tolerance (CVDT)
intended recipient. is a measure of the variation in cell transmission times.
A reliable multicast protocol may ensure reliability Network Related Attributes: The network related
on a per-recipient basis, as well as provide properties
attributes are those that define characteristics of the
that relate the delivery of data to different recipients,
network. The following are some network related
such as e.g. total order, atomicity, or virtual synchrony.
attributes:
Reliable protocols typically incur more overhead
CLR: The Cell loss ratio (CLR) defines the fraction
than unreliable protocols, and as a result, are slower
of cells lost during transmission.
and less scalable. This often is not an issue for unicast
protocols, but it may be a problem for multicast CTD: The Cell transfer delay (CTD) is the average
protocols. time needed for a cell to travel from sources to
TCP, the main protocol used in the Internet today, destination.
is a reliable unicast protocol. CDV: The Cell delay variation (CDV) is the
UDP, often used in computer games or other difference between the CTD maximum and the CTD
situations where speed is an issue and the loss of a little minimum.
data is not as important because of the transitory nature CER: The Cell error ratio (CER) defines the
of the data, is an unreliable protocol. fraction of the cells delivered in error.
Often, a reliable unicast protocol is also connection-
oriented. For example, the TCP/IP protocol is
Q. 1. Explain about TCP flow control.
connection-oriented, with the virtual circuit ID
Ans. TCP utilizes a flow control protocol from one
consisting of source and destination IP addresses and
end to another end for avoiding the sender to send data
port numbers.
too quickly for the TCP receiver to receive and process
QoS (Quality of Service): This quality of service
it in a reliable manner. Machines of diverse network
(QoS) defines a set of attributes related to the
can communicate quickly by having the flow control
performance of the contection. For each conncection
mechanism.
the user can request a particular attribute. Each service
TCP utilizes a sliding window flow control
class is associated with a set of attributes. We can
protocol. In every TCP segment, the receiver is specified
categorize the attributes into those related to the user
in a receive window field, the additional received data
and those related to the network. Figure below shows
that is to buffer for the connection. When a window of
the two categories and some important attributes in each
size 0 is advertised by a receiver, the data sending is
category.
stopped by the sender and starts the persist timer. The
TCP dead lock situation is protected by the persist timer.
The TCP sender sends the packet and the receiver sends
the acknowledgement with the new size of window, at
the time of expiring the persist timer.
Q. 2. Explain IP, TCP and UDP.
Ans. TCP: Transmission control Protocol is used
to establish communication between nodes or networks
User Related Attributes: User related attributes and exchange data packets. It guarantees delivery of
are those attributes that define how fast the user wants data packets in the order they were sent. Hence, it is
to send data. These are negotiated at the time of contract most commonly used in all applications that require
116 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

guaranteed delivery of data. It can handle both timeouts Q. 4. What is Superneting? Explain the concept
(if packets were delayed) and retransmission (if packets of custom Subneting.
were lost). The stream of data is transmitted in segments. Ans. Supernetting or Classless Inter-Domain
The segment header is 32 bit. it is a connectionless Routing (CIDR)
communication protocol at the third level (network) of It is a way to aggregate multiple Internet
the OSI model. addresses of the same class.
IP: Internet protocol is used for transmission of The adjacent network address (e.g.
data over the internet. IP uses IP addresses to identity 192.168.2.0/24 and an address 192.168.3.0/
each machine uniquely. Message is sent using small 24) can be merged (into 192.168.2.0/23) using
packets. The packet contains both the sender and supernetting.
receivers address. IP does not guarantee the delivery in Supernetting is the basis for most routing
the same order as sent. This is because the packets are protocols currently used on the Internet.
sent via different routes. It is a connectionless It is mostly used to combine Class C network
communication protocol at the third level (network) of addresses.
the OSI model. Q. 5. What is UDP protocol?
UDP: User Data Protocol is a communication Ans. User Data Protocol is a communication
protocol. It is normally used as an alternative for TCP/ protocol. It is normally used as an alternative for TCP/
IP. However there are a number of differences between IP. However there are a number of differences between
them. UDP does not divide data into packets. Also, UDP them. UDP does not divide data into packets. Also, UDP
does not send data packets in sequence. Hence, the does not send data packets in sequence. Hence, the
application program must ensure the sequencing. UDP application programme must ensure the sequencing.
uses port numbers to distinguish user requests. It also UDP uses port numbers to distinguish user requests. It
has a checksum capability to verify the data. also has a checksum capability to verify the data.
Q. 3. What is a port and what are port numbers? Q. 6. What is TCP windowing concept?
Ans. Where port numbers are concerned, consider
Ans. TCP windowing concept is primarily used to
an analogy between the computer and the telephone.
avoid congestion in the traffic. It controls the amount
Port numbers and their associated network addresses
of unacknowledged data a sender can send before it gets
work something like telephone numbers and associated
an acknowledgement back from the receiver that it has
extensions. One can think of a network address–say, an
received it.
IP address–like a telephone number. The IP address
Q. 7. What is Trusted and Untrusted Networks?
gives a unique number for reaching someone on the
Ans. Trusted Networks: Such networks allow data
Internet, just as a telephone number allows you to reach
to be transferred transparently. The machines using a
a specific destination on the public telephone network.
trusted network are usually administered by an
(Note that telephone numbers have additional features
that IP addresses do not have, such as area codes that Administrator to ensure that private and secured data is
provide some geographic information about a location.) not leaked. Access to this network is limited. Computers
Many larger organizations set up their telephone using trusted networks are more secured and confidential
networks to use phone extensions. These extensions are because of strong firewalls.
typically 2-5 digit numbers that identify an individual Untrusted Networks: Such networks are usually
phone within the organization. But from the outside, all administered by the owners. They can allow improper
of these extensions are associated with a single telephone access to sensitive or personal data. These machines
number (in the USA, often a 1-800 or 1-888 number). are usually separate. Such machines could me more
A network port number functions similarly to a telephone prone to attacks.
extension. Taken together with a network address, a port Q. 8. What are the different types of VPN?
number identifies both a computer and also a “channel” Ans.
within that computer where network communication will Remote Access VPN: Also called as Virtual
take place. Just as different organizations may use the Private dial-up network (VPDN) is mainly
same extension numbers “inside” their primary phone used in scenarios where remote access to a
number, different computers use the same set of port network becomes essential. Remote access
numbers. VPN allows data to be accessed between a
 TRANSPORT LAYER / 117

company’s private network and remote users Q. 10. What is Tunnelling?

through a third party service provider; Ans. Tunnelling is a mechanism provided to
Enterprise service provider. E.g. Sales team is transfer data securely between two networks. The data
usually present over the globe. Using Remote is split into smaller packets and passed through the
access VPN, the sales updates can be made. tunnel. The data passing through the tunnel has 3 layers
Site to Site VPN - Intranet based: This type of encryption. The data is encapsulated. Tunneling can
of VPN can be used when multiple Remote be approached by Point to Point tunneling protocol.
locations are present and can be made to join Q. 11. What are voluntary and compulsory
to a single network. Machines present on these tunnels?
remote locations work as if they are working Ans. Voluntary Tunnelling: Users computer is an
on a single network. end point of the tunnel and acts as tunnel client. Here,
Site to Site VPN: Extranet based: This type the client or user issues a request to configure and create
of VPN can be used when several different a voluntary tunnel. They require a dial up or LAN
companies need to work in a shared connection. Example of dial up connection is internet
environment. E.g. Distributors and service at home where a call is made to the ISP and connection
companies. This network is more manageable is obtained.
and reliable. Compulsory Tunnelling: In compulsory
Q. 9. What are the different authentication tunnelling, instead of the user a vpn remote access server
methods used in VPNs? configures and creates a tunnel. Hence, the end point is
Ans. The authentication method uses an the Remote sever not the user.
authentication protocol. The methods are: Q. 12. Explain static and dynamic tunnels.
Ans. Tunnels that are created manually are static
EAP Authentication Method: Extensible
tunnels. Tunnels that are auto discovered are dynamic
authentication protocol authenticates remote
tunnels. In dynamic tunnelling, tcp connections can be
access connection. The authentication
checked dynamically. If no connections exist that are
mechanism is decided between the remote
routed through the tunnel, a check for more suitable
VPN client and authenticator (ISA). The
gateway can be done. Static tunnelling may at times
mechanism is typical in which authenticator
require dedicated equipments.
requests for authentication information and the
Q. 13. What is NNTP (Network News Transfer
responses are given by the remote VPN client. Protocol)?
MS Chap Authentication Method: Ans. NNTP or Network News Transfer Protocol is
Microsoft Challenge Handshake Authen- used to manage the notes posted on unset newsgroup (a
tication Protocol (MS-CHAP) starts with the collection of posted notes on a subject posted by
authenticator (Remote access server) different users). NNTP servers are responsible for
challenge. The challenge to the remote access managing Usenet newsgroup collected globally. A NTTP
client sends a session identifier and challenge client is a part of the web browser also called as a news
string. The client in response sends the reader. It uses a reserver port no 119.
nonreversible encryption of the string, the Q. 14. What is POP3 (Post Office Protocol 3)?
identifier and password. Authenticator checks Ans. POP3 or Post Office Box 3 is used fro
the credentials and grants access on a receiving emails. It is a client server protocol which
successful authentication. holds the email. Once the email is downloaded from
Unencrypted Passwords (PAP): Uses plain the server, POP3 deletes it from the server. Ordinal
text passwords. Does not involve encryption. numbers are used to identify specific messages.
Used for less secure clients. Q. 15. What is SNMP (Simple Network
Shiva Password Authentication Protocol Management Protocol)?
(SPAP): It is a password authentication Ans. SNMP or Simple Network Management
protocol. It is less secure as the same user Protocol is typically used for managing the network.
password is always sent in the same reversibly Managing the network includes managing the nodes
encrypted form present in the network. These nodes may be server,
118 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

routers, bridges and hubs. SNMP agents are used to SSRC - Synchronization source identifier uniquely
achieve this. Managing the network is essential because identifies the source of a stream.
it helps to monitor network performance, detect network CSRC–Contributing source IDs enumerate
faults or failures, audit network usage etc. the SNMP contributing sources to a stream which has been
messages like TRAP, GET or SET may be invoked by generated from multiple sources
network elements or network management system. Extension header: - first 32 bit word contains profile
oriented protocol while UDP is connectionless. specific identifier and length specifier
Q. 16. What is RTP? The 32 bits of RTCP header format is as follows:-
Ans. Real-Time Transfer Protocol lays a standard (L to R)
or a way to transfer or manage real time data over a Bit 0-1: Indicates version, currently 2
network. It does not guarantee the delivery of data or Bit 2:- P- indicates padding bytes
provide any quality of service. However, it helps to Bit 3 to 7: Count of number of reception report
manage the data. Which means that RTP can be used blocks
deliver the necessary data to the application to make Bit 8 to 15: Type: Determined RTCP packet type.
sure it can put the received packets in the correct order Type can take values from 0 to 255
Real time data examples could be audio and video. 16 to 31: Length- Length of RTCP packet - 1
Q. 17. What is RTP Multiplexing? SR: Sender Report for transmission and reception
Ans. RTP multiplexing allows multiple media flows from active senders
within a single RTP data payload between two points. RR: Receiver report for reception from in active
This means that it can be used to carry multiple streams senders
of data in one RTP packet. RTP multiplexing will reduce SDES: Source description items
the bandwidth used. RTP multiplexing will also increase BYE- indicates end of participation
scalability. APP: - Application specific functions.
Q. 18. Explain the use of RTP and RTCP Q. 20. Describe how the multicast protocol
protocols. works.
Ans. Use of RTP and RTCP: Ans. Multicast protocol or Internet protocol
delivers a singles message to multiple machines. One
1. RTP can be used to transfer Real time data like
packet from the source is replicated and sent to the
voice packets.
destination. Every multicast message requires a multi-
2. RTP can be used with RTCP which makes it
case group. The group defines the addresses which will
possible to monitor data.
receive the message. The group is defined by the class
3. Packet loss can be detected by RTP using
D address. Different routing protocols are used to find
Sequence number.
the multicast groups and build routes for them. Distance
RTCP provides QoS feedback: Packets lost, round
Vector Multicast protocol is one of them. The receiver,
trip time. to whom the multicast packet is sent to, needs to ‘join’
Q. 19. Describe the format of RTP and RTCP the group. Joining the group is enabled and managed
packets. by IGMP. Multicast routers are used to transmit the
Ans. The 32 bits of RTP packet format is as follows: messages from one network to another.
(L to R) Q. 21. Describe how to control the scope of
Bit 0-1: Indicates version, currently 2 multicast transmissions.
Bit 2: P- indicates padding bytes Ans. Controlling the scope of multicast
Bit 3: X- Indicates presence of extension header transmission restricts the range of group members. TTL
Bit 4 7: CC- Contains number of CSRC identifiers (Time To Live) is one of the mechanisms to limit the
that follows the header scope. If the TTL value is small, packets would only be
Bit 8: M-Current data has some special relevance multicast to smaller distance destinations. More the
(if set) value of TTL, more would be the scope of transmission
Bit 9 15: PT- Indicates format of payload to a larger number of machines. Administrative scoping
Bit 16-31: Sequence number is another mechanism. In this, transmission is restricted
Timestamp: 32bits - time stamp of packet to a specific address space of an organization or a site.
 TRANSPORT LAYER / 119

Q. 22. Explain why use Multicasting. Ans. Stream can be considered as a pipe that allows
Ans.When the same message or packet needs to be full duplex connection. A datagram or a packet on the
sent to multiple destinations, multicasting is used. other hand, has a source and a destination. There is no
Within campus and offices, using multicasting connection. Stream is like a communication channel
file distribution can be done. while datagram is completely self-contained. Streams
System messages, news and videos can be sent provide a reliable and sequenced communication.
at the same time. Datagram’s on the other hand are unreliable and no
More commonly used for audio and video sequence maintained.
streaming. Q. 25. What is a stream socket?
Q. 23. What are Sockets? How do sockets work? Ans. A stream socket provides two way communi-
Ans. A socket is used to connect an application to cations between a client and server. This communication
a network protocol. A socket enables communication is reliable and sequenced. Stream sockets are above TCP
between a client and a server. The communication is to run across any networks. They provide unduplicated
started when the client is assigned a local port number, flow of data and have well established mechanism for
and binds a socket to it. The client writes on the socket creating and destroying connections and for detecting
and gets information from server by reading it. The errors.
Socket class is used to communicate. It provides rich Q. 26. What is SSL?
set of methods for both asynchronous and synchronous Ans. SSL is Secured Socket Layer. SSL is used to
data transfer. ConnectAsynch is used to start an establish a secured and an encrypted connection between
asynchronous connection. SendAsynch and a server and the browser. SSL is most commonly seen
ReceiveAsynch are used to send and receive data. in payment and banking web sites. To create a SSL
Shutdown and close methods are used to shutdown and connection, a SSL certificate needs to be created. If the
close the sockets. website has a SSL certificate installed, a small icon is
Q. 24. Describe difference between datagram displayed in the tool bar to make customers confident
and stream. that the site is secured.
120

Application Layer
11
participant uses a mobile device. Internet domain names
Because the TCP/IP protocol suite was designed are easier to remember than IP addresses such as
before the OSI model, the layers in TCP/IP do not 208.77.188.166 (IPv4) or 2001:db8:1f70::999:
correspond exactly to the OSI layers. TCP/IP layers has de8:7648:6e8 (IPv6). Users take advantage of this when
five layers: The lower four correspond to the lower four they recite meaningful Uniform Resource Locators
OSI layers. TCP/IP application layers, however, is (URLs) and e-mail addresses without having to know
equivalent to the combined session, presentation and how the computer actually locates them.
application layers of the OSI model. This means that The Domain Name System distributes the
all of the functionalities associated with those layers responsibility of assigning domain names and mapping
are handled in one single layer, the application layer. those names to IP addresses by designating authoritative
name servers for each domain. Authoritative name
servers are assigned to be responsible for their particular
DNS domains, and in turn can assign other authoritative name
The Domain Name System (DNS) is a hierarchical servers for their sub-domains. This mechanism has made
distributed naming system for computers, services, or the DNS distributed and fault tolerant and has helped
any resource connected to the Internet or a private
avoid the need for a single central register to be
network. It associates various information with domain
continually consulted and updated.
names assigned to each of the participating entities. Most
In general, the Domain Name System also stores
importantly, it translates domain names meaningful to
other types of information, such as the list of mail servers
humans into the numerical identifiers associated with
that accept email for a given Internet domain. By
networking equipment for the purpose of locating and
providing a worldwide, distributed keyword-based
addressing these devices worldwide.
An often-used analogy to explain the Domain Name redirection service, the Domain Name System is an
System is that it serves as the phone book for the Internet essential component of the functionality of the Internet.
by translating human-friendly computer hostnames into REMOTE LOGGING
IP addresses. For example, the domain name www When a user wants to access an application program
.example.com translates to the addresses 192.0.43.10 or utility located on a remote machine, he or she
(IPv4) and 2620:0:2d0:200::10 (IPv6). performs remote login. Here, the TELNET client and
The Domain Name System makes it possible to server programs came into use. The user sends the
assign domain names to groups of Internet resources keystrokes to the terminal drive where the local
and users in a meaningful way, independent of each operating system accepts the characters but does not
entity’s physical location. Because of this, World Wide interpret them. The characters are sent to the TELNET
Web (WWW) hyperlinks and Internet contact client, which transforms the characters to a universal
information can remain consistent and constant even if character set called network virtual terminal characters
the current Internet routing arrangements change or the and delivers them to the local TCP/IP stack.
 APPLICATION LAYER / 121

FILE TRANSFER assigned. It includes all the “housekeeping”

File transfer is a generic term for the act of that is necessary to keep the network under
transmitting files over a computer network like the control.
Internet. There are numerous ways and protocols to Maintenance is concerned with performing
transfer files over a network. Computers which provide repairs and upgrades–for example, when
a file transfer service are often called file servers. equipment must be replaced, when a router
Depending on the client's perspective the data transfer needs a patch for an operating system image,
is called uploading or downloading. File transfer for when a new switch is added to a network.
the enterprise now increasingly is done with Managed Maintenance also involves corrective and
File Transfer. preventive measures to make the managed
There are two types of file transfers: network run “better”, such as adjusting device
Pull-based file transfers where the receiver configuration parameters.
initiates a file transmission request. Provisioning is concerned with configuring
Push-based file transfers where the sender resources in the network to support a given
initiates a file transmission request. service. For example, this might include setting
A file transfer protocol is a convention that up the network so that a new customer can
describes how to transfer files between two computing receive voice service.
endpoints. They are meant solely to send the stream of A common way of characterizing network
bits stored as a single unit in a file system, plus any management functions is FCAPS–Fault, Configuration,
relevant metadata such as the filename, file size and Accounting, Performance and Security.
timestamp. File transfer protocols usually operate on Functions that are performed as part of network
top of a lower-level protocol in a protocol stack. For management accordingly include controlling, planning,
example, the HTTP protocol operates at the topmost allocating, deploying, coordinating, and monitoring the
application layer of the TCP/IP stack, whereas resources of a network, network planning, frequency
XMODEM, YMODEM, and ZMODEM typically allocation, predetermined traffic routing to support load
operate across RS-232 serial connections. balancing, cryptographic key distribution authorization,
NETWORK MANAGEMENT
configuration management, fault management, security
Network management refers to the activities,
management, performance management, bandwidth
methods, procedures, and tools that pertain to the
management, Route analytics and accounting
operation, administration, maintenance, and
management.
provisioning of networked systems.
Data for network management is collected through
Operation deals with keeping the network (and
the services that the network provides) up and several mechanisms, including agents installed on
running smoothly. It includes monitoring the infrastructure, synthetic monitoring that simulates
network to spot problems as soon as possible, transactions, logs of activity, sniffers and real user
ideally before users are affected. monitoring. In the past network management mainly
Administration deals with keeping track of consisted of monitoring whether devices were up or
resources in the network and how they are down; today performance management has become a
122 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

crucial part of the IT team’s role which brings about a by hyperlinks and URLs, usually accessed by web
host of challenges especially for global organizations. browsers from web servers. In short, the Web can be
CLIENT SERVER APPLICATIONS thought of as an application “running” on the Internet.
A client/server application is a piece of software Viewing a web page on the World Wide Web
that runs on a client computer and makes requests to a normally begins either by typing the URL of the page
remote server. Many such applications are written in into a web browser or by following a hyperlink to that
high-level visual programming languages where UI, page or resource. The web browser then initiates a series
forms, and most business logic reside in the client of communication messages, behind the scenes, in order
application. Often such applications are database to fetch and display it. As an example, consider
applications that make database queries to a remote accessing a page with the URL http://neerajbooks.com/
central database server (this can, however, get much BCA/.
more complicated than that and involve other First, the browser resolves the server-name portion
communication methods) of the URL (neerajbooks.com) into an Internet Protocol
In a database application, data related number address using the globally distributed database known
crunching can occur on the remote database server where as the Domain Name System (DNS); this lookup returns
the processing is close to physical data. An example of an IP address such as 208.80.152.2. The browser then
a database query might be to return the sum of a field requests the resource by sending an HTTP request across
named “dollar amount” where the field name year is the Internet to the computer at that particular address.
“2001”. There may be hundreds of thousands of records It makes the request to a particular application port in
but the client computer does not have to worry about the underlying Internet Protocol Suite so that the
fetching or sorting through all of them itself. The computer receiving the request can distinguish an HTTP
database server will sort through that and just return request from other network protocols it may be servicing
one small record with the result to the client. such as e-mail delivery; the HTTP protocol normally
A client/server application can be cross platform if uses port 80. The content of the HTTP request can be
it is written in a cross platform language, or it can be as simple as the two lines of text
platform specific. In the case of a cross platform WWW PREFIX
language there is an advantage that the application can Many domain names used for the World Wide Web
potentially provide a user interface that is native in begin with www because of the long-standing practice
appearance to the OS or platform environment it is of naming Internet hosts (servers) according to the
running under. services they provide. The hostname for a web server is
An issue of client/server is that the application must often www, in the same way that it may be ftp for an
be installed on each users computer. Depending on the FTP server, and news or nntp for a USENET news
complexity of the program, the environment it is written server. These host names appear as Domain Name
in, and the care the developer took to package the System (DNS) subdomain names, as in www.
program, this can be as easy as creating a shortcut to an example.com. The use of ‘www’ as a subdomain name
executable on a shared network drive or it can be as is not required by any technical or policy standard;
hard as spending hours installing and configuring indeed, the first ever web server was called
runtime software and components on each client nxoc01.cern.ch, and many web sites exist without it.
computer. Many established websites still use ‘www’, or they
WWW invent other subdomain names such as ‘www2’,
The World Wide Web (or the proper World-Wide ‘secure’, etc. Many such web servers are set up such
Web; abbreviated as WWW or W3, and commonly that both the domain root (e.g., example.com) and the
known as the Web is a system of interlinked hypertext www subdomain (e.g. www.example.com) refer to the
documents accessed via the Internet. With a web same site; others require one form or the other, or they
browser, one can view web pages that may contain text, may map to different web sites.
images, videos, and other multimedia and navigate The use of a subdomain name is useful for load
between them via hyperlinks. balancing incoming web traffic by creating a CNAME
The terms Internet and World Wide Web are often record that points to a cluster of web servers. Since,
used in everyday speech without much distinction. currently, only a subdomain can be cname'ed, the same
However, the Internet and the World Wide Web are not result cannot be achieved by using the bare domain root.
one and the same. The Internet is a global system of EMAIL
interconnected computer networks. In contrast, the Web Electronic mail, commonly known as email or e-
is one of the services that runs on the Internet. It is a mail, is a method of exchanging digital messages from
collection of text documents and other resources, linked an author to one or more recipients. Modern email
 APPLICATION LAYER / 123

operates across the Internet or other computer networks.

Some early email systems required that the author and Q. 1. What is HTTP (Hypertext Transfer
the recipient both be online at the same time, in common Protocol)?
with instant messaging. Today’s email systems are based Ans. HTTP or Hyper Text Transfer Protocol is
on a store-and-forward model. Email servers accept, provides a set of rules to transfer files, videos, images
forward, deliver and store messages. Neither the users
over the world wide web. When the web browser is
nor their computers are required to be online
opened, a HTTP request call is made. A web server
simultaneously; they need connect only briefly, typically
contains a HTTP daemon. This daemon is used to wait
to an email server, for as long as it takes to send or
for HTTP requests and handle them when they arrive.
receive messages.
An email message consists of three components, The web browser from where HTTP requests are made
the message envelope, the message header, and the is called as a client. These requests are sent to the server.
message body. The message header contains control It uses a reserved port no 80.
information, including, minimally, an originator’s email Q. 2. What is Application Layer?
address and one or more recipient addresses. Usually Ans. The application layer is located at the top of
descriptive information is also added, such as a subject the TCP/IP protocol layers. This one contains the
header field and a message submission date/time stamp. network applications which make it possible to
Electronic mail predates the inception of the communicate using the lower layers. The software in
Internet, and was in fact a crucial tool in creating it, but this layer therefore communicates using one of the two
the history of modern, global Internet email services protocols of the layer below (the transport layer), i.e.
reaches back to the early ARPANET. Standards for TCP or UDP. In computer networking, an application
encoding email messages were proposed as early as layer firewall is a firewall operating at the application
1973 (RFC 561). Conversion from ARPANET to the layer of a protocol stack. Generally, it is a host using
Internet in the early 1980s produced the core of the various forms of proxy servers to proxy traffic instead
current services. An email sent in the early 1970s looks of routing it. As it works on the application layer, it may
quite similar to a basic text message sent on the Internet inspect the contents of the traffic, blocking what the
today. firewall administrator views as inappropriate content,
MIME such as certain websites, viruses, and attempts to exploit
Multipurpose Internet Mail Extensions (MIME) is known logical flaws in client software, and so forth. An
an Internet standard that extends the format of email to application layer firewall does not route traffic on the
support: network layer. All traffic stops at the firewall which may
Text in character sets other than ASCII
initiate its own connections if the traffic satisfies the
Non-text attachments
rules.
Message bodies with multiple parts
Header information in non-ASCII character Q. 3. Define DNS?
sets Ans. The DNS translates Internet domain and host
MIME’s use, however, has grown beyond names to IP addresses. DNS automatically converts the
describing the content of email to describe content type names we type in our Web browser address bar to the
in general, including for the web (see Internet media IP addresses of Web servers hosting those sites. DNS
type) and as a storage for rich content in some implements a distributed database to store this name
commercial products (e.g. IBM Lotus Domino and IBM and address information for all public hosts on the
Lotus Quickr). Internet.
Virtually all human-written Internet email and a Q. 4. What is a Router?
fairly large proportion of automated email is transmitted Ans. A router is a device or sometimes a software
via SMTP in MIME format. Internet email is so closely in a computer which decides the next network point to
associated with the SMTP and MIME standards that it which a packet should be forwarded to reach its
is sometimes called SMTP/MIME email. destination on Internet. It is usually included as part of
The content types defined by MIME standards are the network switch and is located at a gateway, including
also of importance outside of email, such as in each point-of-presence on the Internet. The router is
communication protocols like HTTP for the World Wide connected to at least two networks and determines which
Web. HTTP requires that data be transmitted in the way to send each information packet based on its
context of email-like messages, although the data most understanding of the state of the networks it is connected
often is not actually email. to the router.
124

Network Applications
12
no clear reason to favour one. The sending agent should
We now come to the layer where all the applications then pick one at random, so it would spread the load
are found. The layer below the application layer are there across multiple mail exchanges for a specific
to provide reliable transport but they do not do real work organization. Finally, the mail destination may be
for user. In this chapter, we will study some real network resolved to multiple different computers, but this time,
applications. ordered by decreasing preference. In this case, if the
However, even in the application layer there is a primary mail exchange is unreachable or busy, a
need for support protocols to allow the applications to secondary computer will accept the message, ensuring
the delivery of the message.
function. Accordingly, we will look at one of these before
CHATTING
starting with the application themselves. After that we
Online chat may refer to any kind of communication
will examine real applications: email, World Wide Web
over the Internet, that offers an instantaneous
and multimedia.
transmission of text-based messages from sender to
receiver, hence, the delay for visual access to the sent
INTERNET APPLICATIONS LIKE EMAILS message shall not hamper the flow of communications
Electronic Mail is the remote messaging system of in any of the directions. Online chat may address as
the Internet. Unlike Telnet or FTP, mail is not meant to well point-to-point communications as well as multicast
be an interactive service. Mail is composed locally and communications from one sender to many receivers.
Online chat in a lesser stringent definition may be
submitted to an agent on the local computer for delivery
primarily any direct text-based one-on-one chat or one-
to a remote computer.
to-many group chat (formally also known as
The protocol used for transmitting messages is
synchronous conferencing), using tools such as instant,
known as the Simple Mail Transfer Protocol, or SMTP.
Internet Relay Chat, talkers and possibly MUDs. The
Like Telnet and FTP, SMTP is designed to be
expression online chat comes from the word chat which
implemented over a reliable service, namely TCP. In a means “informal conversation”.
limited sense, there is error detection in the case of badly Internet Relay Chat (IRC) is a protocol for real-
composed messages, mail addressed to a non-existent time Internet text messaging (chat) or synchronous
user, or undeliverable for some other reason, the conferencing. It is mainly designed for group
originator of the message is sent the infamous bounce communication in discussion forums, called channels,
message, which puts the user in the fault detection/ but also allows one-to-one communication via private
correction loop. message as well as chat and data transfer, including file
To increase the reliability of mail delivery, there sharing.
are a few other mechanisms. If the destination is IRC was created in 1988. Client software is
unreachable, the mail agent will hold the message for a available for every major operating system that supports
set amount of time, usually 3 days, periodically trying Internet access As of April 2011, the top 100 IRC
to deliver the message. If the message is undeliverable networks served more than half a million users at a time,
after the specified time, it is returned to the sender in an with hundreds of thousands of channels operating on a
appropriate bounce message. The mail destination may total of roughly 1,500 servers out of roughly 3,200
also be resolved to multiple different computers, with servers worldwide.
 NETWORK APPLICATIONS / 125

IRC was created by Jarkko Oikarinen in August modes are separate and can use the same letter to mean
1988 to replace a programme called MUT (Multiuser different things (e.g. user mode “i” is invisible mode
Talk) on a BBS called OuluBox in Finland. Oikarinen whilst channel mode “i” is invite only.) Modes are
found inspiration in a chat system known as Bitnet, usually set and unset using the mode command that takes
which operated on the BITNET. a target (user or channel), a set of modes to set (+) or
IRC is an open protocol that uses TCP and, unset (–) and any parameters the modes need.
optionally, TLS. An IRC server can connect to other Some but not all channel modes take parameters
IRC servers to expand the IRC network. Users access and some channel modes apply to a user on a channel
IRC networks by connecting a client to a server. There or add or remove a mask (e.g. a ban mask) from a list
are many client implementations, such as mIRC or associated with the channel rather than applying to the
XChat, and server implementations, e.g. the original channel as a whole. Modes that apply to users on a
IRCd. Most IRC servers do not require users to register channel have an associated symbol that is used to
an account but a user will have to set a nickname before represent the mode in names replies (sent to clients on
being connected. first joining a channel and use of the names command)
IRC was originally a plain text protocol (although and in many clients also used to represent it in the client's
later extended), which on request was assigned port 194/ displayed list of users in a channel or to display an own
TCP by IANA. However, the de facto has always been indicator for a user’s modes.
to run IRC on 6667/TCP and nearby port numbers (for In order to correctly parse incoming mode messages
example TCP ports 6660-6669, 7000) to avoid having and track channel state the client must know which mode
to run the IRCd software with root privileges. is of which type and for the modes that apply to a user
The protocol specified that characters were 8-bit on a channel which symbol goes with which letter. In
but did not specify the character encoding the text was early implementations of IRC this had to be hard-coded
supposed to use. This can cause problems when users in the client but there is now a de-facto standard
using different clients and/or different platforms want extension to the protocol called ISUPPORT that sends
to converse. this information to the client at connect time using
The standard structure of a network of IRC servers numeric 005.
is a tree. Messages are routed along with only necessary There is a small design fault in IRC regarding
branches of the tree but network state is sent to every modes that apply to users on channels: the names
server and there is generally a high degree of implicit message used to establish initial channel state can only
trust between servers. This architecture has a number send one such mode per user on the channel but multiple
of problems. A misbehaving or malicious server can such modes can be set on a single user. For example, if
cause major damage to the network and any changes in a user holds both operator status (+o) and voice status
structure, whether intentional or a result of conditions (+v) on a channel, a new client will be unable to know
on the underlying network, require a net-split and net- the less precedential mode (voice). Work arounds for
join. This results in a lot of network traffic and spurious this are possible on both the client and server side but
quit/join messages to users and temporary loss of none is widely implemented.
communication to users on the splitting servers. Adding Host masks
a server to a large network means a large background A host mask is a unique identifier of an IRC client
bandwidth load on the network and a large memory load connected to an IRC server. IRC servers, IRC services,
on the server. Once established however, each message and other clients including bots can use it to identify a
to multiple recipients is delivered in a fashion similar specific IRC session.
to multicast, meaning each message travels a network The format of a hostmask is nick user@host. The
link exactly once.This is a strength in comparison to host mask looks similar to, but should not be confused
non-multicasting protocols such as Simple Mail Transfer with an e-mail address.
Protocol (SMTP) or Extensible Messaging and Presence The nick part is the nickname chosen by the user
Protocol (XMPP). and may be changed while connected. The user part is
Modes the username reported by indent on the client. If indent
Users and channels may have modes that are is not available on the client, the username specified
represented by single case-sensitive letters and are set when the client connected is used after being prefixed
using the MODE command. User modes and channel with a tilde.
126 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

The host part is the hostname the client is expertise helps us create Internet applications that stay
connecting from. If the IP address of the client cannot up and running around the clock.
be resolved to a valid hostname by the server, it is used Easy-to-maintain: We offer easy-to-use-
instead of the hostname. management solutions and administrative tools enable
Because of the privacy implications of exposing you retain control of your Internet applications. So, you
the IP address or hostname of a client, some IRC no longer have to be completely dependent on a vendor
daemons also provide privacy features, such as to update and make everyday changes to your Web site
InspIRCD or UnrealIRCD’s “+x” mode. This hashes a or Internet application.
client IP address or masks part of a client's hostname, Secure: We design our Internet applications from
making it unreadable to users other than IRCops. Users the start to minimize security threats.
may also have the option of requesting a “virtual host” Digital Saber’s team is competent enough to create
(or “vhost”), to be displayed in the host mask to allow user friendly features that enable our clients to
further anonymity. Some IRC networks such as Free successfully use their Web sites and other Internet
node use these as “cloaks” to indicate that a user is applications.
affiliated with a group or project. Some of the features of our application include the
SOCIAL NETWORKING following:
For scaling up the business ladder it’s not only what
Intelligent data elements
you know, but who you know that counts. And, this is
Multi-tiered navigation systems
where social network applications play a major role.
Syndication and RSS feeds
Social network applications connect people,
Multi-lingual capabilities
information, and ideas and help them to grow and
Global campaign metrics by language and
develop.
bandwidth
We at Digital Saber, design social networking
applications that power Internet business applications. Multimedia integration
So, whether you’re using Internet technology to sell RAIL RESERVATIONS
products, create a social network, manage digital assets, Railway offers industry-leading products and
or automate your internal applications, we offer you the services, consulting and reservation system hosting that
technology that would help you tap new revenue specializes in helping small to medium size business. A
channels and would enable you to reach more customers Railway provides all internet booking solutions and its
at lower costs. features includes Reservation, Inventory Control,
Social Networking applications designed at Digital Custom Tours Booking, Payment Module, Ticketing
Saber offer clients the power to integrate internal Vouchers, and Pick up and Drops as well.
processes along with the power to grow and change. Internet will provide various types of travel based
Building the software infrastructure that powers reservation systems. Our rail reservation system helps
critical Internet applications requires the right set of tools these companies to boost their business throughout the
and the right expertise. Digital Saber possesses a team world. Our all products are very user friendly and easy
of competent and experienced engineers and designers to operate. A railway provides a high performance and
who understand how to translate specific business efficient Internet based booking solution helping your
objectives into a solution that works wonders for the reservation department in a cost-effective way. Railway
clients. to improve the train ticket reservation system, increasing
Cost effective: At Digital Saber, we build Internet customer satisfaction and reducing company’s costs.
engines that help fulfil business objectives and solve INFORMATION SHARING
business problems. We make use of open-source tools The Internet is a global system of interconnected
like Linux, Apache and MySQL, and thus help our computer networks that use the standard Internet
clients save on a lot on money. protocol suite (TCP/IP) to serve billions of users
Scalable: Our solutions are built to grow along with worldwide. It is a network of networks that consists of
your business. Our experts are proficient enough to millions of private, public, academic, business, and
design appropriate solutions for different types and size government networks, of local to global scope, that are
of companies. linked by a broad array of electronic, wireless and optical
Stability is yet another aspect of our applications. networking technologies. The Internet carries an
When downtime is unacceptable, our deep technology extensive range of information resources and services,
 NETWORK APPLICATIONS / 127

such as the inter-linked hypertext documents of the corporation as a result. One example of this practice is
World Wide Web (WWW) and the infrastructure to Microsoft, whose product developers publish their
support email. personal blogs in order to pique the public’s interest in
Many people use the terms Internet and World Wide their work. Collections of personal web pages published
Web, or just the Web, interchangeably, but the two terms by large service providers remain popular, and have
are not synonymous. The World Wide Web is a global become increasingly sophisticated. Whereas operations
set of documents, images and other resources, logically such as Angel fire and Geo Cities have existed since the
interrelated by hyperlinks and referenced with Uniform early days of the Web, newer offerings from, for
Resource Identifiers (URIs). URIs symbolically identify example, Facebook and MySpace currently have large
services, servers, and other databases, and the followings. These operations often brand themselves as
documents and resources that they can provide. social network services rather than simply as web page
Hypertext Transfer Protocol (HTTP) is the main access hosts.
protocol of the World Wide Web, but it is only one of Advertising on popular web pages can be lucrative,
the hundreds of communication protocols used on the and e-commerce or the sale of products and services
Internet. Web services also use HTTP to allow software directly via the Web continues to grow.
systems to communicate in order to share and exchange Email is an important communications service
business logic and data. available on the Internet. The concept of sending
World Wide Web browser software, such as electronic text messages between parties in a way
Microsoft’s Internet Explorer, Mozilla Firefox, Opera, analogous to mailing letters or memos predates the
Apple’s Safari, and Google Chrome, lets users navigate creation of the Internet. Pictures, documents and other
from one web page to another via hyperlinks embedded files are sent as email attachments. Emails can be cc-ed
in the documents. These documents may also contain to multiple email addresses.
any combination of computer data, including graphics, Internet telephony is another common communi-
sounds, text, video, multimedia and interactive content cations service made possible by the creation of the
that runs while the user is interacting with the page. Internet. VoIP stands for Voice-over-Internet Protocol,
Client-side software can include animations, games, referring to the protocol that underlies all Internet
office applications and scientific demonstrations. communication. The idea began in the early 1990s with
Through keyword-driven Internet research using search walkie-talkie-like voice applications for personal
engines likeYahoo! and Google, users worldwide have computers. In recent years many VoIP systems have
easy, instant access to a vast and diverse amount of become as easy to use and as convenient as a normal
online information. Compared to printed media, books, telephone. The benefit is that, as the Internet carries the
encyclopedias and traditional libraries, the World Wide voice traffic, VoIP can be free or cost much less than a
Web has enabled the decentralization of information on traditional telephone call, especially over long distances
a large scale. and especially for those with always-on Internet
The Web has also enabled individuals and connections such as cable or ADSL. VoIP is maturing
organizations to publish ideas and information to a into a competitive alternative to traditional telephone
potentially large audience online at greatly reduced service. Interoperability between different providers has
expense and time delay. Publishing a web page, a blog, improved and the ability to call or receive a call from a
or building a website involves little initial cost and many traditional telephone is available. Simple, inexpensive
cost-free services are available. Publishing and VoIP network adapters are available that eliminate the
maintaining large, professional web sites with attractive, need for a personal computer.
diverse and up-to-date information is still a difficult and Streaming media is the real-time delivery of digital
expensive proposition, however, many individuals and media for the immediate consumption or enjoyment by
some companies and groups use web logs or blogs, end users. Many radio and television broadcasters
which are largely used as easily updatable online diaries. provide Internet feeds of their live audio and video
Some commercial organizations encourage staff to productions. They may also allow time-shift viewing or
communicate advice in their areas of specialization in listening such as Preview, Classic Clips and Listen Again
the hope that visitors will be impressed by the expert features. These providers have been joined by a range
knowledge and free information, and be attracted to the of pure Internet “broadcasters” who never had on-air
128 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

licenses. This means that an Internet-connected device, manage, outsource, or sign a self-funding contract. The
such as a computer or something more specific, can be self-funding model creates portals that pay for
used to access on-line media in much the same way as themselves through convenience fees for certain e-
was previously possible only with a television or radio government transactions, known as self-funding portals.
receiver. The range of available types of content is much Social networking is an emerging area for e-
wider, from specialized technical webcasts to on- democracy. The social networking entry point is within
demand popular multimedia services. Podcasting is a the citizens’ environment and the engagement is on the
variation on this theme, where–usually audio–material citizens’ terms. Proponents of e-government perceive
is downloaded and played back on a computer or shifted government use of social networks as a medium to help
to a portable media player to be listened to on the move. government act more like the public it serves. Examples
These techniques using simple equipment allow can be found at almost every state government portal
anybody, with little censorship or licensing control, to through Facebook,Twitter, and YouTube widgets.
broadcast audio-visual material worldwide. Government and its agents also have the
E-GOVERNANCE opportunity to follow citizens to monitor satisfaction
E-Government (short for electronic government, with services they receive. Through List Servers, RSS
also known as e-gov, digital government, online feeds, mobile messaging, micro-blogging services and
government, or connected government) is digital blogs, government and its agencies can share
interactions between a government and citizens (G2C), information to citizens who share common interests and
government and businesses/Commerce (G2B), concerns.
government and employees (G2E), and also between Whilst e-Government has traditionally been
government and governments /agencies (G2G). understood as being centred around the operations of
‘E-Government’ (or Digital Government) is defined government, e-Governance is understood to extend the
as ‘The employment of the Internet and the world-wide- scope by including citizen engagement and participation
web for delivering government information and services in governance. As such, following in line with the OECD
to the citizens.’ definition of e-Government, e-Governance can be
E-government describes the use of technologies to defined as the use of ICTs as a tool to achieve better
facilitate the operation of government and the governance.
disbursement of government information and services. Advantages
E-government, short for electronic government, deals The ultimate goal of the E-Government is to be
heavily with Internet and non-internet applications to able to offer an increased portfolio of public services to
aid in governments. E-government includes the use of citizens in an efficient and cost effective manner.
electronics in government as large-scale as the use of E-government allows for government transparency.
telephones and fax machines, as well as surveillance Government transparency is important because it allows
systems, tracking systems such as RFID tags, and even the public to be informed about what the government is
the use of television and radios to provide government- working on as well as the policies they are trying to
related information and services to the citizens. implement. Simple tasks may be easier to perform
In the development of public sector or private sector through electronic government access. Many changes,
portals and platforms, a system is created that benefits such as marital status or address changes can be a long
all constituents. Citizens needing to renew their vehicle process and take a lot of paper work for citizens.
registration have a convenient way to accomplish it E-government allows these tasks to be performed
while already engaged in meeting the regulatory efficiently with more convenience to individuals.
inspection requirement. On behalf of a government E-government is an easy way for the public to be more
partner, business provides what has traditionally, and involved in political campaigns. It could increase voter
solely, managed by government and can use this service awareness, which could lead to an increase in citizen
to generate profit or attract new customers. Government participation in elections. It is convenient and cost-
agencies are relieved of the cost and complexity of effective for businesses, and the public benefits by
having to process the transactions. getting easy access to the most current information
To develop these public sector portals or platforms, available without having to spend time, energy and
governments have the choice to internally develop and money to get it.
 NETWORK APPLICATIONS / 129

E-government helps simplify processes and makes software designed for use in business includes IBM
access to government information more easily accessible Lotus Sametime,XMPP and Microsoft Messenger.
for public sector agencies and citizens. For example, One can add friends to a contact or buddy list by
the Indiana Bureau of Motor Vehicles simplified the entering the person’s email address or messenger ID. If
process of certifying driver records to be admitted in the person is online, their name will typically be listed
county court proceedings. Indiana became the first state as available for chat. Clicking on their name will activate
to allow government records to be digitally signed, a chat window with space to write to the other person,
legally certified and delivered electronically by using as well as read their reply.
Electronic Postmark technology. In addition to its Internet Relay Chat (IRC) and other online chat
simplicity, e-democracy services can reduce costs. technologies allow users to join and communicate with
Alabama Department of Conservation & Natural many people at once, publicly. Users may join a pre-
Resources, Wal-Mart and NIC developed an online existing chat room or create a new one about any topic.
hunting and fishing license service utilizing an existing Once inside, you may type messages that everyone else
computer to automate the licensing process. More than in the room can read, as well as respond to from others.
140,000 licenses were purchased at Wal-Mart stores Often there is a steady stream of people entering and
during the first hunting season and the agency estimates leaving. Whether you are in another person’s chat room
it will save $200,000 annually from service. or one you've created yourself, you are generally free
The anticipated benefits of e-government include to invite others online to join you in that room. Instant
efficiency, improved services, better accessibility of messaging facilitates both and interaction.
public services, and more transparency and The goal of groupware software such as Moodle,
accountability. Landing pages, Enterprise Architecture, and sharepoint,
Disadvantages is to allow subjects to share data–such as files, photos,
The main disadvantages concerning e-government text, etc. for the purpose of project work or school work.
is the lack of equality in public access to the internet, The intent is to first form a group and then have them
reliability of information on the web, and hidden agendas collaborate among each other. Clay Shirky defines social
of government groups that could influence and bias software as “software that supports group interaction”.
public opinions. Since groupware supports group interaction (once the
There are many considerations and potential group is formed), it would consider it to be social
implications of implementing and designing e- software.
government, including disintermediation of the Originally modelled after the real-world paradigm
government and its citizens, impacts on economic, of electronic bulletin boards of the world before internet
social, and political factors, vulnerability to cyber was born, internet forums allow users to post a “topic”
attacks, and disturbances to the status quo in these areas. for others to review. Other users can view the topic and
ONLINE PROCESSING post their own comments in a linear fashion, one after
AND COLLABORATIONS the other. Most forums are public, allowing anybody to
Social software applications include communic- sign up at any time. A few are private, gated communities
ation tools and interactive tools. Communication tools where new members must pay a small fee to join, like
typically handle the capturing, storing and presentation the Something Awful Forums.
of communication, usually written but increasingly Forums can contain many different categories in a
including audio and video as well. Interactive tools hierarchy according to topics and subtopics. Other
handle mediated interactions between a pair or group features include the ability to post images or files or to
of users. They focus on establishing and maintaining a quote another user’s post with special formatting in one’s
connection among users, facilitating the mechanics of own post. Forums often grow in popularity until they
conversation and talk. can boast several thousand members posting replies to
An instant messaging application or client allows tens of thousands of topics continuously.
one to communicate with another person over a network There are various standards and claimants for the
in real time, in relative privacy. Popular, consumer- market leaders of each software category. Various add-
oriented clients include AOL Instant Messenger, Google ons may be available, including translation and spelling
speech, ICQ, Meebo, MSN Messenger, Pidgin (formerly correction software, depending on the expertise of the
maig), and Yahoo! Messenger. Instant messaging operators of the bulletin board. In some industry areas,
130 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

the bulletin board has its own commercially successful Design and Implementation Issues
achievements: free and paid hardcopy magazines as well The complexity of groupware development is still
as professional and amateur sites. an issue. One reason for this is the socio-technical
Collaborative software (also referred to as dimension of groupware. Groupware designers do not
groupware) is computer software designed to help only have to address technical issues (as in traditional
people involved in a common task achieve goals. One software development) but also consider the social group
of the earliest definitions of “collaborative software” processes that should be supported with the groupware
is, “intentional group processes plus software to support application. Some examples for issues in groupware
them”. development are:
The design intent of collaborative software Persistence is needed in some sessions. Chat
(groupware) is to transform the way documents and rich and voice communications are routinely non-
media is shared to enable more effective team persistent and evaporate at the end of the
collaboration. Collaboration, with respect to information session. Virtual room and online file cabinets
technology, seems to have several definitions. Some are can persist for years. The designer of the
defensible but others are so broad they lose meaningful collaborative space needs to consider the
application. Understanding the differences in human information duration needs and implement
interactions is necessary to ensure that appropriate accordingly.
technologies are employed to meet interaction needs. Authentication has always been a problem with
Collaboration requires individuals working together groupware. When connections are made point-
in a coordinated fashion, towards a common goal. to-point, of when log-in registration is
Accomplishing the goal is the primary purpose for enforced, it’s clear who is engaged in the
bringing the team together. Collaborative software helps session. However, audio and immoderate
facilitate action-oriented teams working together over sessions carry the risk of unannounced ‘lurkers’
geographic distances by providing tools that aid who observe but do not announce themselves
communication, collaboration and the process of or contribute.
problem solving. Additionally, collaborative software Until recently, bandwidth issues at fixed
may support project management functions, such as task location limited full use of the tools. These are
assignments, time-managing deadlines, and shared exacerbated with mobile devices.
calendars. The artifacts, the tangible evidence of the Multiple input and output streams bring
problem solving process, and the final outcome of the concurrency issues into the groupware
collaborative effort, require documentation and may applications.
involve archiving project plans, deadlines and Motivational issues are important, especially
deliverables. in settings where no pre-defined group process
The use of collaborative software in the workspace was in place.
creates a collaborative working environment (CWE). A Closely related to the motivation aspect is the
collaborative working environment supports people in question of reciprocity. Ellis and others have
both their individual and cooperative work thus giving shown that the distribution of efforts and
birth to a new class of professionals, e-professionals, benefits has to be carefully balanced in order
who can work together irrespective of their geographical to ensure that all required group members
location. really participate.
Finally, collaborative software relates to the notion One approach for addressing these issues is the use
of collaborative work systems which are conceived as of design patterns for groupware design. The patterns
any form of human organization that emerges any time identify recurring groupware design issues and discuss
that collaboration takes place, whether it is formal or design choices in a way that all stakeholders can
informal, intentional or unintentional. Whereas the participate in the groupware development process.
groupware or collaborative software pertains to the Groupware and Levels of Collaboration
technological elements of computer supported Groupware can be divided into three categories
cooperative work, collaborative work systems become depending on the level of collaboration:
a useful analytical tool to understand the behavioural 1. Communication can be thought of as
and organizational variables that are associated to the unstructured interchange of information. A phone call
broader concept of CSCW. or an IM Chat discussion are examples of this.
 NETWORK APPLICATIONS / 131

2. Conferencing (or collaboration level, as it is computers must rely entirely on battery power.
called in the academic papers that discuss these levels) Combined with the compact size of many
refers to interactive work toward a shared goal. mobile devices, this often means unusually
Brainstorming or voting are examples of this. expensive batteries must be used to obtain the
3. Co-ordination refers to complex interde- necessary battery life.
pendent work toward a shared goal. A good metaphor Transmission interferences: Weather, terrain,
for understanding this is to think about a sports team; and the range from the nearest signal point can
everyone has to contribute the right play at the right all interfere with signal reception. Reception
time as well as adjust their play to the unfolding situation in tunnels, some buildings, and rural areas is
– but everyone is doing something different–in order often poor.
for the team to win. That is complex interdependent Potential health hazards: People who use
work toward a shared goal: collaborative management. mobile devices while driving are often
MOBILE APPLICATIONS distracted from driving are thus assumed more
Mobile application development is the process by likely to be involved in traffic accidents.
which application software is developed for small low- (While this may seem obvious, there is
power handheld devices such as personal digital considerable discussion about whether banning
assistants, enterprise digital assistants or mobile phones. mobile device use while driving reduces
These applications are either pre-installed on phones accidents or not.) Cell phones may interfere
during manufacture, downloaded by customers from with sensitive medical devices. There are
various mobile software distribution platforms, or web allegations that cell phone signals may cause
applications delivered over HTTP which use server-side health problems.
or client-side processing (e.g. JavaScript) to provide an Human interface with device: Screens and
“application-like” experience within a Web browser. keyboards tend to be small, which may make
Mobile computing is a form of human-computer them hard to use. Alternate input methods such
interaction by which a computer is expected to be as speech or handwriting recognition require
transported during normal usage. Mobile computing has training.
three aspects: mobile communication, mobile hardware,
and mobile software. The first aspect addresses
communication issues in ad-hoc and infrastructure Q. 1. What is firewall?
networks as well as communication properties, Ans. A firewall is a hardware or software installed
protocols, data formats and concrete technologies. The to provide security to the private networks connected
second aspect is on the hardware, e.g. mobile devices to the internet. They can be implemented in both
or device components. The third aspect deals with the hardware and software, or a combination of both. All
characteristics and requirements of mobile applications. data entering or leaving the Intranet passes through the
Insufficient bandwidth: Mobile Internet firewall which allows only the data meeting the
access is generally slower than direct cable administrators' rules to pass through it.
connections, using technologies such as GPRS Q. 2. What are the types of firewalls?
and EDGE, and more recently HSDPA and Ans. Packet Filtering Firewall: This type of
HSUPA 3G networks. These networks are Firewall detects packets and block unnecessary packets
usually available within range of commercial and makes network traffic release.
cell phone towers. Higher speed wireless Screening Router Firewalls: It’s a software base
LANs are inexpensive but have very limited firewall available in Router provides only light filtering.
range. Computer-based Firewall: It’s a firewall stored
Security standards: When working mobile, in server with an existing Operating System like
one is dependent on public networks, requiring Windows and UNIX.
careful use of VPN. Security is a major concern Hardware-base Firewall: Its device like box
while concerning the mobile computing allows strong security from public network. Mostly used
standards on the fleet. One can easily attack by big networks.
the VPN through a huge number of networks Proxy Server: Proxy server allows all clients to
interconnected through the line. access Internet with different access limits. Proxy server
Power consumption: When a power outlet or has its own firewall which filters the all packet from
portable generator is not available, mobile web server.
132

13
NETWORK DESIGN AND SECURITY

Building a Simple Network

less) that do nothing but fulfil DHCP services. That’s a
A computer is a programmable machine designed waste. While technology professionals should always
to automatically carry out a sequence of arithmetic or work to deploy the best technologies for clients or
logical operations. The particular sequence of operations organisations in which they work, IT pros have a
can be changed readily, allowing the computer to solve responsibility to ensure the correct technologies are
more than one kind of problem. An important class of deployed.
computer operations on some computing platforms is If a company employs fewer than 10 employees
the accepting of input from human operators and the and isn’t forecasting aggressive growth, it’s likely a peer-
output of results formatted for human consumption. The to-peer network of Windows XP Professional systems
interface between the computer and the human operator will meet the organization’s needs. Should the
is known as the user interface. organization grow, the investment in XP Pro systems
Conventionally, a computer consists of some form can still be leveraged. When exceeding 12 or so
of memory, at least one element that carries out employees, though, organizations are typically better
arithmetic and logic operations, and a sequencing and off with a server.
control unit that can change the order of operations based One easy but all-too-common mistake can be easily
on the information that is stored. Peripheral devices avoided; never deploy Windows XP Home systems in
allow information to be entered from an external source, office or business workgroup environments. The
and allow the results of operations to be sent out. consumer OS doesn’t offer the security controls
Windows XP Professional boasts, nor can Windows XP
Home join a Small Business Server-powered domain
EXAMPLES OF DESIGNING THE DEVELOPING (should the organization grow to the point that it requires
SMALL NETWORKS centralized administration).
When deploying Windows Small Business Server, 2. Good Topology is Planned, not Accidental:
or any small business network, most implementations Poorly configured networks are everywhere. The
can be optimized if you keep the following 10 elements number of three- and four-hop networks I encounter as
in mind: an IT consultant is mind numbing. Adding four- and
1. Servers are not Always Necessary: AMI five-port routers/switches here and there isn’t the proper
Research reports that of 68.5 million small businesses way to connect networks. Although it’s an easy
worldwide, more than 21 million have multiple PCs but approach, performance inevitably becomes a problem.
no server. That figure can be found in Microsoft Thus, it comes as no surprise when a small business
promotional materials designed to help IT consultants with nine systems and four switches approaches me
understand the small business space and sell servers. complaining of poor network performance. Worse yet,
But the fact is, many organisations don’t have servers many of these networks often possess multiple
because they don’t need them. Other organizations have 192.168.x.y and 10.0.x.y subnets.
servers because an IT professional or consultant sold Replacing three four-port switches with a single 16-
them one, but they don’t actually require it. port switch typically introduces dramatic performance
Customers whose previous IT consultants sold them improvements. Even with only 10 or 12 employees, the
US$4,000 servers (with dual-Intel Pentium CPUs, no additional ports often prove handy for providing a single
 BUILDING A SIMPLE NETWORK / 133

interface for all client systems, a server if it’s present, three different subnets, or worse, encountering multiple
and firewall and DSL or T1 connectivity. DHCP devices serving up the same IP addresses within
It’s easy to fall into the trap of simply adding the same ranges (don’t laugh, it happens), always plot
switches as a network expands, but before you add a network topologies and the corresponding IP subnets
second switch, plot the network on paper. Count the on paper. A world of mistakes (and hurt) can be easily
number of hops a connection must navigate before avoided, as discrepancies are easily spotted when a
reaching the gateway. Often, you’ll find investing in a network is properly documented on paper or within
few extra cable drops (to avoid requiring an additional Visio.
switch in a quickly expanding office) provides Multiple subnets aren’t always bad, of course.
performance gains that more than justify the expense. Occasionally, a small business may require two subnets.
Whenever you prepare to add new systems or are When security is of particular concern, maintaining
prepping a new network, consider its topology carefully. sensitive data on a secondary subnet available only to
Work to minimize the number of hops data must travel, limited personnel (and typically removed from wireless
even if investing in a switch with more ports or new
connectivity) may prove best. Such secondary subnets
cable drops is required.
also prove helpful when you want to limit VPN or remote
3. Network Equipment Deserves Investment:
The average lifespan of a network switch or firewall is traffic to specific network segments.
probably four to five years. Others may tell you three 5. You should Choose Internal Domain Names
years is the average, but let’s be honest -- most Carefully: The Microsoft community recommends
organizations try to maximise their investments by rolling out servers using the .local domain. The problem
running equipment longer. is that Macintosh systems encounter trouble resolving
If you try to save funds by purchasing cheaper addresses with Active Directory when the local domain
network equipment, you’ll almost certainly pay the price is used because Apple’s own Rendezvous technology
down the road. Less expensive switches, firewalls, VPN was designed to use the local namespace.
appliances, and routers are available, sure. But they’re If an organization doesn’t need to include Macs on
made with lesser quality materials. That means the its network and never will, the problem isn’t likely to
power supplies are less robust. Ports are more likely to be an issue. But if Macs are added in the future, selecting
fail. Cheap fans are more likely to malfunction, thereby
a different namespace will help avoid having to make
resulting in failure due to overheating.
other changes to enable the Apple systems to properly
When designing or “re-architecting” a small
resolve DNS requests.
business network, budget funds appropriately for
commercial-class network equipment. Buy the best the That said, you should also guard against using
organization can afford. Although there are no moving publicly routed domain names as an internal domain
parts, and the equipment is likely parked in a closet namespace.User logons sometimes took 12 to 15
where no one will ever see it to appreciate it, most minutes to complete. Design networks to use top-level
aspects of an organisation’s data and communications domains that aren’t publicly routed on the Internet.
will run through the equipment. If there’s ever an Many administrators prefer the local or lan top-
element that justifies purchasing quality components, level domains. For testing purposes, the test domain
that’s it. works well. The example domain, meanwhile, is a safe,
4. IP Addressing Deserves Attention: Just as a if unsavoury, bet.
network’s topology deserves attention and planning, so 6. Data Requires Segregation: When designing a
too does a network’s IP addressing scheme. The small business network, plan on separating the network’s
popularity of universal threat management (UTM)
data to maximize data protection, backups, and recovery.
appliances and proprietary router and firewall operating
How? Follow the best practice of installing an
systems, such as those found in Cisco, SonicWALL, and
other companies’ devices, often introduces a variety of operating system’s files on a different partition (and
operating subnets. preferably a different hard disk) than that housing user
As a result, troubleshooting connection failures, and application data. Better yet, ensure a Windows
performance issues, and other problems is made server’s Exchange database is parked on its own
exponentially more difficult. Instead of maintaining partition or disk, too.
134 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Ultimately, that’s a lot of partitions and disks. Most CALs if you wish to tie licensing to individual users;
small businesses are unlikely to maintain the three (or select device CALs if you wish to tie Microsoft licensing
six, if disk mirroring is implemented) hard disks such to servers and PCs (such as when you have multiple
planning requires, but at least give it some thought. shifts of call centre or customer service personnel using
Organisations unable to maintain (due to cost the same PCs throughout a typical day).
restrictions) separate hard disks for an operating SBS ships with five user CALs and you can buy
system’s files, user data, and the Exchange database additional CALs in sets of fives and tens. But once you
should place that much more emphasis on making sure hit 75 users, you’re done. If you anticipate exceeding
that backup operations properly complete on a regular 75 users anytime in the foreseeable future, you should
basis. begin making plans to migrate to a standard version of
7. Wireless Technologies are Problematic: the full-blown Windows server platform instead.
Although they’re full of promise, wireless networks 9. Terminal Services Changed in SBS 2003:
frequently prove frustrating, introducing problems like Terminal Services licensing changed with the
security vulnerabilities and flaky network connections. introduction of SBS 2003. Small Business Server 2000
From relatively weak WPA keys to easily defeated MAC supported Terminal Services running in Application
filtering, wireless security (or the lack thereof) is Mode, but SBS 2003 does not. What’s that mean?
infamous. Tack on reduced throughput, the need to If users must access network applications powered
position antenna and access points carefully, and the by the server, organizations should design their small
potential for introducing yet another routing device business network to use Windows Server 2003 instead.
serving potentially conflicting IP addresses, and you may Only users possessing administrator rights can log on
be prompted to rethink whether wireless access is really remotely to a Windows SBS 2003 desktop and even
required on a network. then, SBS 2003 supports a limit of two such concurrent
Certainly, occasions will arrive in which an sessions.
organisation’s users require the mobility wireless Microsoft claims Terminal Services in Application
networking provides. Or a business may occupy a Mode was disabled in SBS 2003 as a security
facility in which running required Ethernet cables simply precaution. As mentioned previously, however, the
isn’t viable. When designing or planning such networks, feature is still available in Windows Server 2003.
be sure to seek routing equipment that also includes Ultimately, your best bet when designing a small
wireless service. By combining routing/firewall/VPN business network is to limit as many remote connections
features and wireless connectivity in a single device, as possible. But should your organization require such
some of the routing issues that arise when adding connectivity, be aware that SBS 2003 may not meet your
multiple devices to a network can be eliminated requirements.
(although you’re still stuck with many of wireless’ 10. Review Features-Versus-Needs Before
security headaches). Purchase and Deployment: This most important step
8. License Planning is Critical: Planning peer- is often forgotten. Planning discussions almost always
to-peer workgroup licensing requirements is simple. begin with an energetic needs assessment. But
Purchase as many systems running Windows XP technology professionals servicing smaller
Professional as you require and roll them out. organizations often leave such planning meetings to
Planning Small Business Server licensing begin ordering equipment and coordinating a resulting
requirements is another matter. Windows SBS ships with migration or rollout.
just five client access licenses (CALs). Unlike other Once discussions of IP addressing schemes, Active
Windows server platforms, the user CALs associated Directory configuration, Exchange administration,
with Windows Small Business Server cover all SBS' performance reporting, and the like begin, it’s easy for
technologies, including Exchange and SQL. technology professionals to lose focus and forget
Although that’s relatively simple, there’s still some business requirements.
confusion regarding the two types of SBS CALs For example, business owners may be focused on
available: User CALs and Device CALs. Choose User adding the ability for travelling employees to securely
 BUILDING A SIMPLE NETWORK / 135

access sensitive resources while on the road via VPN. For all practical purposes, you can work with these
Or they may be eagerly awaiting the opportunity to three CATx types in the same manner. When there is a
integrate user administration, file and print sharing, difference Itll point that out. The two important ‘rules’
e-mail, and Web hosting on a single system in-house. to remember are:
If the technology professional doesn’t design the Do not remove any more of the cables jacket
small business network to power the services the than you absolutely have to.
business demands, the table is set for disappointment. The un-twisted part of the wires CANNOT
It’s easy to forget to design the network to support secure exceed 1/2”.
VPN connections or to specify the proper IP address These rules are intended to limit how much you
necessary for Exchange e-mail and Web hosting once can untwist the twisted pairs. So lets move on.
you become wrapped up in the many other details that Configuring the Hardware: So what do you need
inevitably accompany such a project. and how do you hook everything up? Here are examples
When designing a small business network, between of several set ups.
discussing the organization’s needs and announcing the
resulting solution, technology professionals would be
wise to schedule an extra meeting. The purpose? To
ensure the proposed solution and new network design
not only possess the capacity and features required to
meet the organization’s business requirements, but to
confirm that the new network will be implemented in a
manner that maximizes its capabilities and enables users
This is about as simple a configuration as you can
to meet the organization's business requirements.
get. With this you can have all of your PC networked
STRUCTURE CABLING together and enjoy all of the benefits that affords you
CATx is a term use to describe CAT5 and CAT5e Basic Network with 2 Hubs or Switches: This is
and CAT6. CATx cables have 8 conductors grouped as still a pretty basic configuration. It shows how to use
four twisted pairs. It’s the twisted pairs that are key to multiple Hubs or Switches to give you more ports.
the ability of these cables to carry such high speeds
digital signals (100 Mbps and 1000 Mbps) over such a
long distance of 100 metres. Some things common to
CAT5, CAT5e and CAT6 cable are–they all have four
twisted pairs and they all use the same color wires. See
the CAT5 and CAT5e cables in the photo below. Notice
how the pairs consist of a solid coloured insulated wire
paired with a white coloured insulated wire that has a
stripe of the same colour as the solid wire. Note here
too how CAT5 and CAT5e look so similar. CAT6 photos
aren’t shown here. In fact not to cover CAT6 cable or
it’s connectors and terminating procedures. 1st likely
be adding info on CAT6 soon.
Network with Broadband at the CWP and 2
Hubs or Switches: This configuration adds Broadband
access. If you have broadband access in any form such
as high speed cable, DSL or whatever you’ll need some
form of Broadband Modem. Most of these modems
come as a stand alone device or as a plug-in PC card.
For a networked home, the stand alone device makes
136 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

the most logical configuration because it will operate IP address. Without a Router most ISP’s will charge
independently of any one PC. Locate this broadband you extra for each PC that is sharing their broadband
modem at the CWP. Now there are many manufacturers access. The router will allow you to avoid these extra
of these Hubs, Switches and Modem devices and they charges.
will likely label their input/output connections
differently than It have in these drawings. These
drawings just show a generic configuration.

Network with Broadband at the CWP and a

Router/Switch and 2 Hubs or Switches: This
configuration adds an ALL IMPORTANT Broadband
INTEGRATING HOME COMPUTERS
Router/Switch to your Broadband access. This
AND DEVICES
Broadband Router/Switch adds an important security
Before beginning to assemble your new computer,
feature to your network - a Hardware Firewall. This
gather everything you need (your computer tool kit,
Router will also allow several of your PC’s to share one
components and manuals) together in one place.
Place something over your table to avoid damaging the surface. An anti-static mat is best because it will not
only protect your table from scratches, but also protect your computer’s components from static damage.

Installing the Motherboard in your Homebuilt Computer: Installing the motherboard in your homebuilt
computer usually is pretty easy (though sometimes knuckle-busting). Basically, you just line up the board with the
mounting holes and rear-panel openings, and then screw it in. (If you purchased a barebones computer kit the
motherboard may already be installed; in which case you can skip this page if you like.)
 BUILDING A SIMPLE NETWORK / 137

prevent improper insertion, but they’re not all the same.

Sometimes you have to look at the pins, and sometimes
you have to line up an arrow. Whatever the case, look
carefully before you insert the processor to make sure
that you're inserting it correctly. If your processor came
with a manual, read it.
Mounting the CPU Cooler: The CPU cooler
assembly is positioned over the processor and is usually
secured to the motherboard by metal clips that hold the
Attach the Power Connectors: Finally, connect heat sink down tightly against the processor.
the ATX power connector from the power supply to the
motherboard. Do this now. If you forget about it and
later fire up your computer while the ATX connector is
not connected to anything, then you will fry your
computer’s power supply.

Notice that on bail-type sockets, both the socket

and the heat sink are offset from center a little to
accommodate the bail. Make sure you install the cooler
with its offset on the same side as the offset on the socket.
Use a slotted screwdriver to gently, but firmly, hook
Installing the Processor in Your Homebuilt the retaining clips under the tabs on the processor socket.
Computer: The microprocessor is often both the most Be very careful not to let the screwdriver slip. If it
delicate and the most expensive part of a homebuilt scratches the surface of the motherboard, the mobo could
computer. As such, it deserves and requires special care. be ruined.
Handle processors gingerly, and never touch the pins Some cooling assemblies use plastic clips that
or conductors with your fingers. simply slide straight down over the processor and snap
Inserting the Processor Correctly: Processor into place. These types of assemblies usually don’t
sockets are keyed to prevent improper installation. But require any tools to install.
because they're so delicate and the pins so fine, it’s really, Proper Insertion of the RAM Modules: RAM
really easy to permanently damage them if you’re not modules are keyed with little notches that fit over
careful. corresponding tabs in the RAM slots. We’ve circled
them in the picture on the right (click the picture for a
close-up).

Notice in the picture on the right that both the

processor and the socket have two corners without pins.
This is to prevent the processor from being installed Before inserting the RAM module, make sure that
the wrong way. All processors have some system to the notches and tabs are correctly lined up. If there are
138 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

more notches then there are tabs (or vice-versa), then manage shared resources. So before installing PCI cards,
you have the wrong RAM for your motherboard. Same check the motherboard and expansion card manuals for
thing goes for if the notches and tabs don’t match up. any recommendations for slot assignments; and if one
You have the wrong RAM, and all the force in the world or more of your cards don’t work (or if they cause system
won’t make it the right RAM. instability), try changing the slots before you trash the
Proper Insertion of Expansion Cards: Like cards. Sometimes that’s all it takes.
RAM, expansion cards and slots are keyed. They have Once you have determined which cards will be
little notches with corresponding tabs in the slot that installed in which slots, actually installing them is
are designed to prevent you from installing the wrong simple. Place the computer on its side so the slots on
card. the motherboard face up, align the card in the slot
perpendicular to the motherboard (that is, straight up,
because the computer is on its side), and push down
until you feel the card “pop” into place.

So if the card doesn’t seem to fit, check those

notches and tabs. Don't break out a hammer and try to
pound it in. You probably are trying to insert the wrong
kind of card (or insert the card in the wrong kind of
If the card doesn’t seat itself using fingertip
slot).
Notice in the picture on the right that the AGP video pressure, place your palm over the card and push down
slot is set back from the rest of the slots and is of a firmly and evenly until you feel the card pop into place.
different size. In addition, the various ridges, tabs, and But first check the slots, notches, and tabs to make sure
so forth on the card and the slot are intended to help you’re not trying to install the card in the wrong slot.
prevent incorrect insertion or incompatible cards. You Finally, secure the card into place by screwing the
should read the documentation for your motherboard card’s metal bracket into the screw hole over the
and cards to make sure they are compatible. expansion slot opening on the back of the case. Some
The card shown in this picture is a network card cases don’t use screws, and instead have some sort of
that fits into the PCI slot. Notice that it is keyed to the metal or plastic clip that holds the card (or all of them,
slot. (Also note that the card is only slanted to make it
sometimes) in the motherboard.
easier for you to see the slot. Expansion cards, like
RAM, are pushed straight down into their slots, like in
the next picture down.)

Connecting SATA Drives: If you are using a SATA

(Serial ATA) drive, you don’t have to worry about
It usually doesn’t matter which PCI cards are jumper settings or clumsy ribbon cables. The connectors
installed in which slots. But sometimes it does, on the motherboard will look like those in the picture
depending on how a particular motherboard and OS on the right.
 BUILDING A SIMPLE NETWORK / 139

the USB header on the motherboard, or to an add-on

card.
Installing the Cabinet Fan: While we’re at it, let’s
also install the cabinet fan. On our case, the cabinet fan
is mounted in a fan shroud, which is in turn snapped
into the computer case. Other cases simply have holes
for the fan to be directly mounted to the case using
special screws that come with the fan.

The first drive on the primary SATA controller

should be connected to the system hard drive. Other
than that, it really doesn’t make much of a difference
which connectors you use for the other drives.
Once again, the cable should be inserted straight
down into the connector. Don’t force it! If it doesn’t
seem to want to go in, make sure that the cable is
properly oriented on the connector. SATA connectors
are keyed to prevent improper insertion, so if it doesn’t
fit easily, you’re probably trying to attach it backwards.
If you force it, you'll break it, and your expensive new
motherboard will be ruined.
Connecting the Power Cables: Finally, attach the
power connectors to the drives.
Make sure that the power connectors are attached
using the correct polarity. The sockets and connectors
are shaped so that they should only fit the correct way Some people always mount the cabinet fans to blow
unless you force them. So don’t force them. If it doesn’t the air outward, to avoid sucking dust into the computer.
fit without forcing, then you probably are trying to Others say you should install the front panel fan to draw
connect it backwards! the air inward, and the rear panel fan to blow the air
outward. We say it depends. If the computer is going to
be used in a dusty place, point the fan to blow the air
out to avoid dust. If not, then point it in to increase
airflow.
Time to Double-Check: Before firing up your new
computer, take a few moments to double check the
following items:
Check all the fans to make sure they are properly
connected. Starting up your computer with the
Installing Auxiliary Drives in your Computer: CPU fan disconnected will likely kill your
There are several different types of auxiliary drives processor!
available for your homebuilt computer, such as CD-RW Make sure that all wires and cables are safely
and DVD-RW drives, ZIP drives, tape drives, and drive- tied away from the fans. Neatness counts. Use
like devices such as card readers. The installation
plastic cable ties, not metal twist-ties. If you
procedures for all of these devices are quite similar.
can’t get plastic cable ties, then use electrical
For this demonstration, we will be installing a CD-
tape.
RW drive on an EIDE interface. Other types of interfaces
Check that all of the power and data cables are
also are available (SATA and SCSI). The process is
securely connected and are attached in the
exactly the same for installing a device like a card reader,
except that there may be an additional connection to correct polarity.
140 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Make sure that there are no tools, screws, or Make sure your ICS host computer has an active
jumpers floating around in the case. Internet connection. To establish your Internet
Check that all expansion cards and RAM connection, run the New Connection Wizard.
modules are securely seated. Run the Windows XP professional Network
Setup Wizard on the ICS host computer.
Run the Network Setup Wizard on the other
computers on your network.

Q. 1. Define some objectives creation of small

business/operational networks?
Ans. The network is designed to achieve several
specific business/operational objectives:
1. Secure Service: The main objective of this
CREATING A SMALL NETWORKING network is to provide secure administrative computing
Create a home or small business network to harness service to the State Office and two districts. It is designed
the power of all your computer resources, for work and to be functionally and physically isolated from access
play. Follow step-by-step instructions for planning and by people not employed by the Maryland public
setting up a secure and reliable home or small business education system so as to minimize the risk of
network. Save yourself time and money–read the section unauthorized use.
on Hardware Requirements for Home Networking to 2. Integration and Update: Presently there are
double-check your shopping list before visiting your many LANs in the Maryland public education system,
local computer store. but much of the equipment is out of date, many of the
Sketch out your network: draw a diagram of LANs are incompatible with each other, and not
your house or office where each computer and connected in a system-wide network. This proposal
printer is located. Or, you can create a table that describes a WAN that integrates and updates these LANs
lists the hardware on each computer. to support productive collaboration across the system.
Next to each computer, note the hardware, such 3.Versatile Information Processing: The network
as modems and network adapters, each will enable users to retrieve, process, and store ASCII
computer has. and non-ASCII text, still graphics, audio, and video from
Choose your Internet Connection Sharing (ICS) any connected computer.
host computer. It is recommended that this 4. Collaboration: The network will combine the
computer be running Windows XP Home power and capabilities of diverse equipment across the
Edition or Windows XP Professional and have state to provide a collaborative medium that helps users
a working Internet connection. combine their skills regardless of their physical location.
Determine the type of network adapters you A network for this educational community will enable
need for your home or small office network: people to share information and ideas easily so they
Ethernet, home phone line network adapter can work more efficiently and productively.
(HPNA), wireless, or IEEE 1394. 5. Scalability: The design is scalable so that more
Make a list of hardware you need to purchase. district offices can be added as funding becomes
This includes modems, network adapters, hubs, available without having to redo the installed network.
and cables. Q. 2. What is a network? What are the different
Purchase the hardware. kinds of network? Explain them.
Install the network adapters and modems to Ans. A network is a group of computers or nodes
create your network connections on each connected together. They are connected with each other
computer. by communication paths.
Physically connect the computers together. Plug Types of Networks
in the cables to hubs, phone jacks, and the LAN: Local Area Network connects a group of
computer. nodes covering a small physical area. LAN’s are most
Turn on all computers and printers. commonly seen in offices, building etc. LAN’s enable
 BUILDING A SIMPLE NETWORK / 141

higher transfer rate of data, smaller coverage of area of 2 and so on. The host decrements the TTL value by
and hence less wiring. one and the packet is forwarded to the next host, when
WAN: Wide Area Network connects a group of a packet passes through a host.
nodes covering a wide area. WAN typically connects When a packet (with a time-to-live value of 1)
and allow communication between regions or national reaches a host, the packets will be discarded by the host,
boundaries. The most common example of WAN is and an ICMP time exceeded packet is sent to the sender.
internet. A list of hosts, whose packets have traversed enrouted
VPN: Virtual Private Network connects or links to the destination, is utilized by the trace route utility.
nodes in some larger area by open connections or virtual Q. 4. What is TCP connection establishment and
circuits in some larger network (e.g. the Internet) instead tear down?
of by physical wires. It is used for secure communication Ans. The following is the process for TCP
through the public internet. VPN alone may not support connection:
explicit security features, such as authentication or 1. As connection is made by synchronizing sender
content encryption.
and receiver and the OS is informed about the
Intranet: It is a set of networks under the control
connection establishment.
of a single administrative person. It can be considered
2. Sender starts sending information and gets
as an internal network of an organization. If it is large,
acknowledged. Soon after the sender sends the data, a
web servers are used to provide information to the users.
Extranet: It is a network that restricts itself within timer is started.
a single organization. It can be categorized as WAN, 3. When the sender is not received any acknow-
MAN etc. however; it cannot have a single LAN. It must ledgement ever after exceeding the timer’s limit, the
have a connection (at least one) with external network. data is transmitted.
Q.3. Explain the working of trace route. 4. If windowing occurs, buffer on the receiver is
Ans. Trace route works by increasing the time-to- full, then a stop signal is sent by the sender and stops
live value of every successive packets sent in batch. sending information.
Every three packets that were sent first, have a time-to- 5. Soon after processing all data, a go signal is sent
live value 1. The next three packets have a TTL value by the sender, followed by transmitting data again.
142

Introduction to Network Architectures

14
protocol to X.25, ISO 8208, is compatible with X.25,
The network layer is concerned with getting packets but additionally includes provision for two X.25 DTEs
from the source all the way to the distinction. Getting to to be directly connected to each other with no network
the destination may require making many haps at in between. By separating the Packet-Layer Protocol,
intermediate routers along the way. This function clearly ISO 8208 permits operation over additional networks
contrasts with that of the data link layers, which has the such as ISO 8802 LLC2 (ISO LAN) and the OSI data
more modest goal of just moving frames from one end link layer.
of a wire to the other. Thus, the network layer is the X.25 originally defined three basic protocol levels
lowest layer that deals with end-to-end transmission. or architectural layers. In the original specifications
To achieve its goals, the network layer must know these were referred to as levels and also had a level
about the topology of the communication subnet and number, whereas all ITU-T X.25 recommendations and
choose appropriate paths through it. Four important ISO 8208 standards released after 1984 refer to them
functionality of the network layers are: as layers. The layer numbers were dropped to avoid
(i) Store and forward packet switching confusion with the OSI Model layers.
(ii) Service provided to the transport layer. Physical layer: This layer specifies the
(iii) Connection less services. physical, electrical, functional and procedural
(iv) Connection oriented services. characteristics to control the physical link
between a DTE and a DCE. Common impleme-
ntations use X.21, EIA-232, EIA-449 or other
X.25 serial protocols.
X.25 is an ITU-T standard protocol suite for packet Data link layer: The data link layer consists of
switched wide area network (WAN) communication. An the link access procedure for data interchange
X.25 WAN consists of packet-switching exchange (PSE) on the link between a DTE and a DCE. In its
nodes as the networking hardware, and leased lines, implementation, the Link Access Procedure,
Plain old telephone service connections or ISDN Balanced (LAPB) is a data link protocol that
connections as physical links. X.25 is a family of manages a communication session and controls
protocols that was popular during the 1980s with the packet framing. It is a bit-oriented protocol
telecommunications companies and in financial that provides error correction and orderly
transaction systems such as automated teller machines. delivery.
The X.25 specification defines only the interface Packet layer: This layer defined a packet-layer
between a subscriber (DTE) and an X.25 network protocol for exchanging control and user data
(DCE). X.75, a very similar protocol to X.25, defines packets to form a packet-switching network
the interface between two X.25 networks to allow based on virtual calls, according to the Packet
connections to traverse two or more networks. X.25 does Layer Protocol.
not specify how the network operates internally-many The X.25 model was based on the traditional
X.25 network implementations used something very telephony concept of establishing reliable circuits
similar to X.25 or X.75 internally, but others used quite through a shared network, but using software to create
different protocols internally. The ISO equivalent “virtual calls” through the network. These calls
 INTRODUCTION TO NETWORK ARCHITECTURES / 143

interconnect “data terminal equipment” (DTE) Error Control: Error recovery procedures at the
providing endpoints to users, which looked like point- packet layer assume that the data link layer is responsible
to-point connections. Each endpoint can establish many for retransmitting data received in error. Packet layer
separate virtual calls to different endpoints. error handling focuses on resynchronizing the
X.25 was developed in the era of dumb terminals information flow in calls, as well as clearing calls that
connecting to host computers, although it also can be have gone into unrecoverable states:
used for communications between computers. Instead Level 3 Reset packets, which re-initializes the
of dialing directly “into” the host computer–which flow on a virtual call (but does not break the
would require the host to have its own pool of modems virtual call).
and phone lines, and require non-local callers to make Restart packet, which clears down all virtual
long-distance calls–the host could have an X.25 calls on the data link and resets all permanent
connection to a network service provider. Now dumb- virtual circuits on the data link.
terminal users could dial into the network’s local “PAD” FRAME RELAY
(Packet Assembly/Disassembly facility), a gateway Frame relay is a telecommunication service
device connecting modems and serial lines to the X.25 designed for cost-efficient data transmission for
link as defined by the X.29 and X.3 standards. intermittent traffic between local area networks (LANs)
Having connected to the PAD, the dumb-terminal and between end-points in a wide area network (WAN).
user tells the PAD which host to connect to, by giving a Frame relay puts data in a variable-size unit called a
phone-number-like address in the X.121 address format frame and leaves any necessary error correction
(or by giving a host name, if the service provider allows (retransmission of data) up to the end-points, which
for names that map to X.121 addresses). The PAD then speeds up overall data transmission. For most services,
places an X.25 call to the host, establishing a virtual the network provides a permanent virtual circuit (PVC),
call. Note that X.25 provides for virtual calls, so appears which means that the customer sees a continuous,
to be a circuit switched network, even though in fact dedicated connection without having to pay for a full-
the data itself is packet switched internally, similar to time leased line, while the service provider figures out
the way TCP provides connections even though the
the route each frame travels to its destination and can
underlying data is packet switched. Two X.25 hosts
charge based on usage. An enterprise can select a level
could, of course, call one another directly; no PAD is
of service quality–prioritizing some frames and making
involved in this case. In theory, it doesn’t matter whether
others less important. Frame relay is offered by a number
the X.25 caller and X.25 destination are both connected
of service providers, including AT&T. Frame relay is
to the same carrier, but in practice it was not always
provided on fractional T-1 or full T-carrier system
possible to make calls from one carrier to another.
carriers. Frame relay complements and provides a mid-
For the purpose of flow-control, a sliding window
range service between ISDN, which offers bandwidth
protocol is used with the default window size of 2. The
at 128 Kbps, and Asynchronous Transfer Mode (ATM),
acknowledgements may have either local or end to end
which operates in somewhat similar fashion to frame
significance. A D bit (Data Delivery bit) in each data
packet indicates if the sender requires end to end relay but at speeds from 155.520 Mbps or 622.080
acknowledgement. When D = 1, it means that the Mbps.
acknowledgement has end to end significance and must Frame relay is based on the older X.25 packet-
take place only after the remote DTE has acknowledged switching technology which was designed for
receipt of the data. When D = 0, the network is permitted transmitting analog data such as voice conversations.
(but not required) to acknowledge before the remote Unlike X.25 which was designed for analog signals,
DTE has acknowledged or even received the data. frame relay is a fast packet technology, which means
While the PAD function defined by X.28 and X.29 that the protocol does not attempt to correct errors. When
specifically supported asynchronous character terminals, an error is detected in a frame, it is simply “dropped.”
PAD equivalents were developed to support a wide (thrown away). The end points are responsible for
range of proprietary intelligent communications devices, detecting and retransmitting dropped frames. (However,
such as those for IBM System Network Architecture the incidence of error in digital networks is
(SNA). extraordinarily small relative to analog networks.)
144 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Frame relay is often used to connect local area Public telephone operators (PTOs) own and build
networks with major backbones as well as on public networks of the first two types and provide services to
wide area networks and also in private network the public under license from the national government.
environments with leased lines over T-1 lines. It requires Virtual Network Operators (VNOs) lease capacity
a dedicated connection during the transmission period. wholesale from the PTOs and sell on telephony service
It’s not ideally suited for voice or video transmission, to the public directly.
which requires a steady flow of transmissions. However, The public switched telephone network (PSTN) is
under certain circumstances, it is used for voice and the network of the world's public circuit-switched
video transmission. telephone networks. It consists of telephone lines, fibre
Frame relay relays packets at the Data Link layer optic cables, microwave transmission links, cellular
of the Open Systems Interconnection (OSI) model rather networks, communications satellites, and undersea
than at the Network layer. A frame can incorporate telephone cables, all inter-connected by switching
packets from different protocols such as Ethernet and centres, thus allowing any telephone in the world to
X.25. It is variable in size and can be as large as a communicate with any other. Originally a network of
thousand bytes or more. fixed-line analog telephone systems, the PSTN is now
Advantages of Frame Relay: Frame Relay offers almost entirely digital in its core and includes mobile
an attractive alternative to both dedicated lines and X.25 as well as fixed telephones.
networks for connecting LANs to bridges and routers. The technical operation of the PSTN utilizes
The success of the Frame Relay protocol is based on standards created by the ITU-T. These standards allow
the following two underlying factors: different networks in different countries to interconnect
Because virtual circuits consume bandwidth seamlessly. There is also a single global address space
only when they transport data, many virtual for telephone numbers based on the E.163 and E.164
circuits can exist simultaneously across a given standards. The combination of the interconnected
transmission line. In addition, each device can networks and the single numbering plan make it possible
use more of the bandwidth as necessary, and for any phone in the world to dial any other phone.
thus operate at higher speeds. ATM NETWORK
The improved reliability of communication Asynchronous Transfer Mode (ATM) is a standard
lines and increased error-handling sophistication switching technique designed to unify telecommuni-
at end stations allows the Frame Relay protocol cation and computer networks. It uses asynchronous
to discard erroneous frames and thus eliminate time-division multiplexing, and it encodes data into
time-consuming error-handling processing. small, fixed-sized cells. This differs from approaches
TELEPHONE NETWORK such as the Internet Protocol or Ethernet that use variable
A telephone network is a telecommunications sized packets or frames. ATM provides data link layer
network used for telephone calls between two or more services that run over a wide range of OSI physical Layer
parties. links. ATM has functional similarity with both circuit
There are a number of different types of telephone switched networking and small packet switched
network: networking. It was designed for a network that must
A fixed line network where the telephones must handle both traditional high-throughput data traffic (e.g.
be directly wired into a single telephone file transfers), and real-time, low-latency content such
exchange. This is known as the public switched as voice and video. ATM uses a connection-oriented
telephone network or PSTN. model in which a virtual circuit must be established
A wireless network where the telephones are between two endpoints before the actual data exchange
mobile and can move around anywhere within begins. ATM is a core protocol used over the SONET/
the coverage area. SDH backbone of the public switched telephone network
A private network where a closed group of (PSTN) and Integrated Services Digital Network
telephones are connected primarily to each other (ISDN), but its use is declining in favour of All IP.
and use a gateway to reach the outside world. At this rate, a typical full-length 1500 byte (12000-
This is usually used inside companies and call bit) data packet would take 77.42 µs to transmit. In a
centres and is called a private branch exchange lower-speed link, such as a 1.544 Mbit/s T1 line, a 1500
(PBX). byte packet would take up to 7.8 milliseconds.
 INTRODUCTION TO NETWORK ARCHITECTURES / 145

A queuing delay induced by several such data (permanent virtual circuits or PVCs) or paths (permanent
packets might exceed the figure of 7.8 ms several times virtual paths or PVPs) require that the circuit is
over, in addition to any packet generation delay in the composed of a series of segments, one for each pair of
shorter speech packet. This was clearly unacceptable interfaces through which it passes.
for speech traffic, which needs to have low jitter in the PVPs and PVCs, though conceptually simple,
data stream being fed into the codec if it is to produce require significant effort in large networks. They also
good-quality sound. A packet voice system can produce do not support the re-routing of service in the event of a
this low jitter in a number of ways: failure. Dynamically built PVPs (soft PVPs or SPVPs)
Have a playback buffer between the network and PVCs (soft PVCs or SPVCs), in contrast, are built
and the codec, one large enough to tide the codec by specifying the characteristics of the circuit (the
over almost all the jitter in the data. This allows service “contract”) and the two end points.
smoothing out the jitter, but the delay introduced Finally, ATM networks create and remove switched
by passage through the buffer would require virtual circuits (SVCs) on demand when requested by
echo cancellers even in local networks; this was an end piece of equipment. One application for SVCs
considered too expensive at the time. Also, it is to carry individual telephone calls when a network of
would have increased the delay across the telephone switches are inter-connected using ATM.
channel, and conversation is difficult over high- SVCs were also used in attempts to replace local area
delay channels. networks with ATM.
Build a system that can inherently provide low ISP
jitter (and minimal overall delay) to traffic that An ISP (Internet service provider) is a company
needs it. that provides individuals and other companies access
Operate on a 1:1 user basis (i.e. a dedicated to the Internet and other related services such as Web
pipe). site building and virtual hosting. An ISP has the
ATM operates as a channel-based transport layer, equipment and the telecommunication line access
using virtual circuits (VCs). This is encompassed in the required to have a point-of-presence on the Internet for
concept of the Virtual Paths (VP) and Virtual Channels. the geographic area served. The larger ISPs have their
Every ATM cell has an 8- or 12-bit Virtual Path Identifier own high-speed leased lines so that they are less
(VPI) and 16-bit Virtual Channel Identifier (VCI) pair dependent on the telecommunication providers and can
defined in its header. Together, these identify the virtual provide better service to their customers. Among the
circuit used by the connection. The length of the VPI largest national and regional ISPs are AT&T WorldNet,
varies according to whether the cell is sent on the user- IBM Global Network, MCI, Netcom, UUNet, and
network interface (on the edge of the network), or if it PSINet.
is sent on the network-network interface (inside the ISPs also include regional providers such as New
network). England’s NEARNet and the San Francisco Bay area
As these cells traverse an ATM network, switching BARNet. They also include thousands of local
takes place by changing the VPI/VCI values (label providers. In addition, Internet users can also get access
swapping). Although the VPI/VCI values are not through online service providers (OSP) such as America
necessarily consistent from one end of the connection Online and Compuserve.
to the other, the concept of a circuit is consistent (unlike The larger ISPs interconnect with each other
IP, where any given packet could get to its destination through MAE (ISP switching centers run by MCI
by a different route than the others. WorldCom) or similar centers. The arrangements they
Another advantage of the use of virtual circuits make to exchange traffic are known as peering
comes with the ability to use them as a multiplexing agreements. There are several very comprehensive lists
layer, allowing different services (such as voice, Frame of ISPs world-wide available on the Web.
Relay, n* 64 channels, IP). The VPI is useful for An ISP is also sometimes referred to as an IAP
reducing the switching table of some virtual circuits (Internet access provider). ISP is sometimes used as an
which have common paths. abbreviation for independent service provider to
ATM can build virtual circuits and virtual paths distinguish a service provider that is an independent,
either statically or dynamically. Static circuits separate company from a telephone company.
146 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

IPV4 AND IPV6 OVERVIEW (though the bytes are still present in the actual network
Internet Protocol (IP) technology was developed address) as follows:
in the 1970s to support some of the first research E3D7::51F4:9BC8:C0A8:6420
computer networks. Today, IP has become a worldwide Finally, many IPv6 addresses are extensions of IPv4
standard for home and business networking as well. Our addresses. In these cases, the rightmost four bytes of an
network routers, Web browsers, email programs, instant IPv6 address (the rightmost two byte pairs) may be
messaging software–all rely on IP or other network rewritten in the IPv4 notation. Converting the above
protocols layered on top of IP. example to mixed notation yields
Two versions of IP technology exist today. E3D7::51F4:9BC8:192.168.100.32
Essentially all home computer networks use IP version IPv6 addresses may be written in any of the full,
4 (IPv4), but an increasing number of educational and shorthand or mixed notation illustrated above.
research institutions have adopted the next generation IPv4 Address Classes
IP version 6 (IPv6). The IPv4 address space can be subdivided into 5
IPv4 Addressing Notation classes–Class A, B, C, D and E. Each class consists of a
An IPv4 address consists of four bytes (32 bits). contiguous subset of the overall IPv4 address range.
These bytes are also known as octets. With a few special exceptions explained further
For readability purposes, humans typically work
below, the values of the leftmost four bits of an IPv4
with IP addresses in a notation called dotted decimal.
address determine its class as follows:
This notation places periods between each of the four
Class A - 0nnnnnnn hhhhhhhhhhhhhhhhhhhhhhhh
numbers (octets) that comprise an IP address. For
First bit 0; 7 network bits; 24 host bits
example, an IP address that computers see as
Initial byte: 0 – 127
00001010 00000000 00000000 00000001
is written in dotted decimal as 126 Class As exist (0 and 127 are reserved)
10.0.0.1 16,777, 214 hosts on each Class A
Because each byte contains 8 bits, each octet in an Class B – 10nnnnnnnnnnnnnn hhhhhhhhhhhhhhhh
IP address ranges in value from a minimum of 0 to a First two bits 10; 14 network bits; 16 host bits
maximum of 255. Therefore, the full range of IP Initial byte: 128 – 191
addresses is from 0.0.0.0 through 255.255.255.255. That 16,384 Class Bs exist
represents a total of 4,294,967,296 possible IP addreses. 65,532 hosts on each Class B
IPv6 Addressing Notation Class C – 110nnnnnnnnnnnnnnnnnnnnn hhhhhhhh
IP addresses change significantly with IPv6. IPv6 First three bits 110; 21 network bits; 8 host bits
addresses are 16 bytes (128 bits) long rather than four Initial byte: 192 – 223
bytes (32 bits). This larger size means that IPv6 supports 2,097,152 Class Cs exist
more than 254 hosts on each Class C
300,000,000,000,000,000,000,000,000,000,000,000,000 Class D-1110mmmmmmmmmmmmmmmmmm
possible addresses! In the coming years, as an increasing mmmmmmmmmm
number of cell phones, PDAs, and other consumer First four bits 1110; 28 multicast address bits
electronics expand their networking capability, the Initial byte: 224 – 247
smaller IPv4 address space will likely run out and IPv6 Class Ds are multicast addresses
address become necessary. Class E - 1111rrrr rrrrrrrr rrrrrrrr rrrrrrrr
IPv6 addresses are generally written in the First four bits 1111; 28 reserved address bits
following form: Initial byte: 248 - 255
hhhh:hhhh:hhhh:hhhh:hhhh:hhhh:hhhh:hhhh Reserved for experimental use.
In this full notation, pairs of IPv6 bytes are
separated by a colon and each byte in turns is represented
as a pair of hexadecimal numbers, like in the following Q. 1. Explain the concepts and capabilities of IP
example: Datagram.
E3D7:0000:0000:0000:51F4:9BC8:C0A8:6420 Ans. Data transmitted over an internet using IP is
As shown above, IPv6 addresses commonly contain carried in messages called IP datagrams.
many bytes with a zero value. Shorthand notation in Like all network protocol messages, IP uses a
IPv6 removes these values from the text representation specific format for its datagrams.
 INTRODUCTION TO NETWORK ARCHITECTURES / 147

The IPv4 consists of the following fields: time is 25.6 microseconds, what is the minimum size
Version: the version of IP used to generate the of the frame?
datagram. For IPv4, this is of course the number Ans. The frame transmission time is Tfr = 2 X Tp
4. = 51.2 microseconds.
Internet Header Length: Specifies the length of This means, in the worst case, a station needs to
the IP header, in 32-bit words. transmit for a period of 51.2 microseconds to detect the
Type of service: information to provide quality collision.
of service features. The minimum size of the frame is 10 Mbps X 51.2
Total Length. microseconds = 512 bits or 64 bytes.
Identification: It can be used if the datagram Q. 4. What are the two important TCP Socket
must be fragmented by a router during delivery classes? Explain them.
and then for reassembling. Ans. The two most important socket classes are:
Flags: (Don’t Fragment, More Fragment) Socket Class: The Socket class provides
Fragment Offset: When fragmentation of a enriched methods and properties for network
message occurs, this field specifies the offset, communications. It allows both synchronous
or position.
and asynchronous data transfer. Shutdown
Time to Live: specifies how long the datagram
method should be used to end the Socket once
is allowed to “live”.
the data transfer is complete. Using
Protocol: ICMP, TCP, UDP, etc.
SetSocketOption the socket can be configured.
Header Checksum: to provide basic protection
Declaration:
against corruption in transmission.
Public Class Socket _
Source Address.
Destination Address. Implements IDisposable
Options. Server Socket Class: The server Socket class
Padding: if number of bits used for options is is used to implement server sockets. Server
not a multiple of 32, then paddin is added. sockets are used to respond to requests received
Data over the network.
Q 2. Given the minimum size of an Ethernet Q. 5. What are the advantages and disad-
frame as 72 bytes and the speed of electromagnetic vantages of Java Sockets?
Propagation in thick coaxial cable as 60 percent of Ans. Advantages:
the speed of light (3´ 108 metres/sec) Calculate the Flexible and powerful.
maximum cable length for the 10mbps Ethernet Cause low network traffic if efficiently used.
collision domain. Only updated information can be sent.
Ans. The minimum size of the Ethernet frame is Disadvantages:
72 bytes. The Java applets can establish communication
To send 576 bits in 10 Base 5, it takes (576 bits / only with the machine requested and not with
10 Mbps ) = 57.6 microseconds. any other machine on the network.
Before the last bit is sent, the first bit must have Sockets allow only raw data to be sent. This
reached the last node and if there is a collision, it must means that both client and server need to have
be sensed by the sender during this time. That means, mechanisms to interpret the data.
the round trip time is 57.6 micro seconds. Q. 6. What is race condition? How does the race
The speed of propagation = 3 X 108 X (60 / 100) condition occur?
metres/sec Ans. A race condition occurs when there are two
= 180x106 metres/sec or more processes trying to access the same data
Therefore the maximum cable length concurrently or performing the same operation
= (57.6/2) X 10–6 X 180 X 106 metres concurrently. Output of the operation depends on the
= 180 X 57.6 / 2 metres sequence of requests of the operations.
= 5184 metres In networking, a race condition occurs when two
Q 3. A network using CSMA/CD has a or more try to access the same channel. This usually
bandwidth of 10 Mbps. If the maximum propagation occurs in networks having lag times.
148 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

In a computer memory, a race condition occurs the protocol are (from bottom to top): Physical layer,
when a request to read and write data is received at the Data link layer, Network layer, transport layer and
same time. application layer, also called as the OSI model. In TCP/
In logic gates, a race condition occurs when some IP , IP is responsible for forwarding packets while TCP
inputs encounter conflicts. ensures the correct delivery of data from client to server.
Q. 7. What is meant by Symmetric TCP detects loss of data as well.
Multiprocessing (SMP)? Q. 9. What is difference between baseband and
Ans. Symmetric Multiprocessing allows any broadband transmission ?
processor to do work of another processor. SMP allows Ans. In a baseband transmission, the entire
multiple processors to process programs that are a part bandwidth of the cable is consumed by a single signal.
of common OS and memory. Most common example In broadband transmission, signals are sent on multiple
of SMP is OLTP where many users access the same
frequencies, allowing multiple signals to be sent
database in a relatively simple set of transactions. SMP
simultaneously.
efficiently balances the workload of the processors.
Q. 8. What is TCP / IP protocol? Q. 10. What is the difference between routable
Ans. Transmission Control Protocol/Internet and non-routable protocols?
Protocol: It is a family of protocols used for Ans. Routable protocols can work with a router
communication and connection between hosts on the and can be used to build large networks. Non-Routable
internet. It is the most widely used standard for protocols are designed to work on small, local networks
transmitting data over the internet. The four layers in and cannot be used with a router.
 149

Introduction to Wireless and Mobile Networks

15
few meters for television remote control, or as far as
Wireless communications is one of the most active thousands or even millions of kilometers for deep-space
areas of technology development of our time. This radio communications. It encompasses various types of
development is being driven primarily by the fixed, mobile, and portable two-way radios, cellular
transformation of what has been largely a medium for telephones, personal digital assistants (PDAs), and
supporting voice telephony into a medium for supporting wireless networking. Other examples of wireless
other services, such as the transmission of video, images, technology include GPS units, Garage door openers or
text, and data. Thus, similar to the developments in wire garage doors, wireless computer mice, keyboards and
line capacity in the 1990s, the demand for new wireless Headset (telephone/computer), headphones, radio
capacity is growing at a very rapid pace. Although there receivers, satellite television, broadcast television and
are, of course, still a great many technical problems to cordless telephones.
be solved in wireline communications, demands for Wireless communications can be via:
additional wireline capacity can be fulfilled largely with radio frequency communication,
the addition of new private infrastructure, such as microwave communication, for example long-
additional optical fibre, routers, switches, and so on. range line-of-sight via highly directional
On the other hand, the traditional resources that antennas, or short-range communication,
have been used to add capacity to wireless systems are infrared (IR) short-range communication, for
radio bandwidth and transmitter power. Unfortunately, example, from consumer IR devices such as
these two resources are among the most severely limited remote controls or via Infrared Data
in the deployment of modern wireless networks: radio Association (IrDA).
bandwidth because of the very tight situation with regard Mobile telephones: One of the best-known
to useful radio spectrum, and transmitter power because examples of wireless technology is the mobile phone,
mobile and other portable services require the use of also known as a cellular phone, with more than 4.6
battery power, which is limited. These two resources billion mobile cellular subscriptions worldwide as of
are simply not growing or improving at rates that can the end of 2010. Started in 1983, these wireless phones
support anticipated demands for wireless capacity. On use radio waves to enable their users to make phone
the other hand, one resource that is growing at a very calls from many locations worldwide. They can be used
rapid rate is that of processing power. within range of the mobile telephone siteused to house
the equipment required to transmit and receive the radio
signals from these instruments.
INTRODUCTION TO WIRELESS MODERN WIRELESS COMMUNICATION
COMMUNICATION SYSTEMS SYSTEMS AND GENERATIONS
Wireless telecommunications is the transfer of Wireless data communications are an essential
information between two or more points that are not component of mobile computing. The various available
physically connected. Distances can be short, such as a technologies differ in local availability, coverage range
150 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

and performance, and in some circumstances, users must when it was revealed that Microsoft’s implementation
be able to employ multiple connection types and switch of encryption in some of its 27 MHz models was highly
between them. To simplify the experience for the user, insecure.
connection manager software can be used, or a mobile
VPN deployed to handle the multiple connections as a Global Roaming
More Capacity,
secure, single virtual network. Supporting technologies High Speed Data
include:
Wi-Fi is a wireless local area network that enables
portable computing devices to connect easily to the
Internet. Standardized as IEEE 802.11, Wi-Fi
approaches speeds of some types of wired Ethernet. Wi-
Fi has become the de facto standard for access in private 1G 2G 2.5G 3G

homes, within offices, and at public hotspots. Some

These standards are set by the International
businesses charge customers a monthly fee for service,
Telecommunication Union or ITU for short. It is a
while others have begun offering it for free in an effort
specialized agency of the United Nations, founded in
to increase the sales of their goods.
1865, which is responsible for information and
Cellular data service offers effective coverage
communication technologies. The ITU co-ordinates the
within a range of 10-15 miles from the nearest cell site.
shared global use of the radio spectrum, works to
Speeds have increased as technologies have evolved,
improve telecommunication infrastructure and
from earlier technologies such as GSM, CDMA and
establishes related worldwide telecommunication
GPRS, to 3G networks such as W-CDMA, EDGE or
standards.
CDMA2000.
It all started with 1G, the first generation of wireless
Mobile Satellite Communications may be used
telephone technology and mobile telecommunications.
where other wireless connections are unavailable, such
Introduced sometime in the 1980s, 1G networks used
as in largely rural areas or remote locations. Satellite
analog signals, as opposed to digital signals used by all
communications are especially important for
the successive generations of mobile technologies. In
transportation, aviation, maritime and military use.
1G network, voice calls were simply modulated to a
Wireless Energy Transfer: Wireless energy
higher frequency, typically to 150MHz and up. The first
transfer is a process whereby electrical energy is
commercially available cellular network using 1G
transmitted from a power source to an electrical load
standard was introduced by NTT (Nippon Telegraph
that does not have a built-in power source, without the
and Telephone) in 1979 in Japan. NTT is also a
use of interconnecting wires.
shareholder and technology provider for TATA
Computer Interface Devices: Answering the call
DoCoMo in India.
of customers frustrated with cord clutter, many
The second generation, 2G, was commercially
manufactures of computer peripherals turned to wireless
launched for the GSM standard in 1991 by Radiolinja,
technology to satisfy their consumer base. Originally
currently known as Elisa Oyj, in Finland. 2G allowed
these units used bulky, highly limited transceivers to
for enhanced data services and also introduced short
mediate between a computer and a keyboard and mouse,
messaging service (SMS). Since the introduction of 2G,
however, more recent generations have used small, high
voice communications were digitally encrypted. This
quality devices, some even incorporating Bluetooth.
allowed for greater privacy, efficient data transfer and
These systems have become so ubiquitous that some
also less expensive equipment. Two revisions or
users have begun complaining about a lack of wired
additions to this generation are sometimes referred to
peripherals. Wireless devices tend to have a slightly
2.5G and 2.75G. The combined introduction of GPRS
slower response time than their wired counterparts,
(General Packet Radio Services) and the usage of
however the gap is decreasing. Concerns about the
CDMAone networks collectively came to be known as
security of wireless keyboards arose at the end of 2007,
 INTRODUCTION TO WIRELESS AND MOBILE NETWORKS / 151

2.5G. GPRS provided data transfer rates from 56-115 In addition, we can send Data much faster that that
kbit/s. So, services like WAP (Wireless Application of the previous generations. A 4G cellular system must
Protocol) and MMS (Multimedia Messaging) were have target peak data rates of up to approximately 100
introduced, along with Internet services. 2.75G was the Mbit/s for high mobility such as mobile access and up
name given to the evolution of EDGE (Enhanced Data to approximately 1 Gbit/s for low mobility such as
rates for GSM Evolution) or Enhanced GPRS (EGPRS). nomadic/local wireless access, according to the ITU
This was due to the introduction of 8PSK encoding, requirements. Scalable bandwidths up to at least 40 MHz
which facilitated higher data transfer rates of up to should be provided. A 4G system is expected to provide
236.8kbits/s, almost triple of the previous rates. This is a comprehensive and secure all-IP based solution where
the most widely used generation of mobile telecommuni- facilities such as IP telephony, ultra-broadband Internet
cation networks, and is what most mobile phones in access, gaming services and HDTV streamed
India are now using. multimedia may be provided to users. Although legacy
The third generation, 3G, was introduced by NTT systems are in place to adopt existing users, the
DoCoMo in Japan, in 2001. Although initially limited infrastructure for 4G will be only packet-based (all-IP).
in scope, it was a leap forward. 3G used completely Some proposals suggest having an open Internet
different radio frequencies from 2G, so it required platform. Technologies considered to be early 4G
different equipment to achieve the new high data transfer include: Flash-OFDM, the 802.16e mobile version of
rates. Also, the enormous costs of additional spectrum WiMax (also known as WiBro in South Korea), and
licensing fees delayed the introduction of 3G in many HC-SDMA.
countries. 3G data transfer rates are 384 kbits/s to 2 Indian market still awaits the full implementation
Mbits/s, so it allows for previously unavailable services of 3G and there are lot of issues relating to its availability.
like video calls, video conferencing, online conference When 4G comes along it will also be seen whether it
call, mobile TV, online gaming etc. These speeds are succeeds in the Indian context or not.
broadband equivalent, so the applications and INTRODUCTION TO CELLULAR
capabilities are enhanced greatly. Along with these MOBILE SYSTEMS
services, 3G provides greater security and privacy. As Cellular is one of the fastest growing and most
with 2G, minor evolution of the standards resulted in demanding telecommunications applications. Today, it
3.5G and 3.75G. Again, these standards allowed for represents a continuously increasing percentage of all
higher data transfer rates, exceeding 2 Mbits/s, reaching new telephone subscriptions around the world. Currently
about 14 Mbits/s. there are more than 45 million cellular subscribers
The next generation, 4G mobile phones are all worldwide, and nearly 50 per cent of those subscribers
set to provide data transfer rates of 100 Mbit/s to 1 are located in the United States. It is forecasted that
Gbit/s, which is mind boggling, to say the least. Such cellular systems using a digital technology will become
speeds are not even present in wired networks the universal method of telecommunications. By the year
commercially. 2005, forecasters predict that there will be more than
3G has just been launched in India and is available 100 million cellular subscribers worldwide. It has even
on select mobile operators for select cities. But to access been estimated that some countries may have more
these services, a 3G compatible mobile phone is mobile phones than fixed phones by the year 2000.
required. The concept of cellular service is the use of low-
The Forth Generation could be ready for power transmitters where frequencies can be reused
implementation around 2012. within a geographic area. The idea of cell-based mobile
Then with the case of Fourth Generation that is 4G radio service was formulated in the United States at Bell
in addition to that of the services of 3G have some Labs in the early 1970s. However, the Nordic countries
additional features such as Multi-Media Newspapers. were the first to introduce cellular services for
You can also watch T.V. programmes with the clarity commercial use with the introduction of the Nordic
that of an ordinary T.V. Mobile Telephone (NMT) in 1981.
152 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Cellular systems began in the United States with Figure 2 shows a metropolitan area configured as a
the release of the advanced mobile phone service traditional mobile telephone network with one high-
(AMPS) system in 1983. The AMPS standard was power transmitter.
adopted by Asia, Latin America, and Oceanic countries, 2. Mobile Telephone System Using the Cellular
creating the largest potential market in the world for Concept: Interference problems caused by mobile units
cellular. using the same channel in adjacent areas proved that all
In the early 1980s, most mobile telephone systems channels could not be reused in every cell. Areas had to
were analog rather than digital, like today’s newer be skipped before the same channel could be reused.
systems. One challenge facing analog systems was the Even though this affected the efficiency of the original
inability to handle the growing capacity needs in a cost- concept, frequency reuse was still a viable solution to
efficient manner. As a result, digital technology was the problems of mobile telephony systems.
welcomed. The advantages of digital systems over Engineers discovered that the interference effects
analog systems include ease of signaling, lower levels were not due to the distance between areas, but to the
of interference, integration of transmission and ratio of the distance between areas to the transmitter
switching, and increased ability to meet capacity power (radius) of the areas. By reducing the radius of
demands. an area by 50 per cent, service providers could increase
1. Mobile Communications Principles: Each the number of potential customers in an area fourfold.
mobile uses a separate, temporary radio channel to talk Systems based on areas with a one-kilometre radius
to the cell site. The cell site talks to many mobiles at would have one hundred times more channels than
once, using one channel per mobile. Channels use a pair systems with areas 10 kilometres in radius. Speculation
of frequencies for communication one frequency (the led to the conclusion that by reducing the radius of areas
forward link) for transmitting from the cell site and one to a few hundred meters, millions of calls could be
frequency (the reverse link) for the cell site to receive served.
calls from the users. Radio energy dissipates over The cellular concept employs variable low-power
distance, so mobiles must stay near the base station to levels, which allow cells to be sized according to the
maintain communications. The basic structure of mobile subscriber density and demand of a given area. As the
networks includes telephone systems and radio services. population grows, cells can be added to accommodate
Where mobile radio service operates in a closed network that growth. Frequencies used in one cell cluster can be
and has no access to the telephone system, mobile reused in other cells. Conversations can be handed off
telephone service allows interconnection to the from cell to cell to maintain constant phone service as
telephone network. the user moves between cells.
Early Mobile Telephone System Architecture: The cellular radio equipment (base station) can
Traditional mobile service was structured in a fashion communicate with mobiles as long as they are within
similar to television broadcasting: One very powerful range. Radio energy dissipates over distance, so the
transmitter located at the highest spot in an area would mobiles must be within the operating range of the base
broadcast in a radius of up to 50 kilometres. The cellular station. Like the early mobile radio system, the base
concept structured the mobile telephone network in a station communicates with mobiles via a channel. The
different way. Instead of using one powerful transmitter, channel is made of two frequencies, one for transmitting
many low-power transmitters were placed throughout a to the base station and one to receive information from
coverage area. For example, by dividing a metropolitan the base station.
region into one hundred different areas (cells) with low- 3. Cellular System Architecture: Increases in
power transmitters using 12 conversations (channels) demand and the poor quality of existing service led
each, the system capacity theoretically could be mobile service providers to research ways to improve
increased from 12 conversations or voice channels using the quality of service and to support more users in their
one powerful transmitter to 1,200 conversations systems. Because the amount of frequency spectrum
(channels) using one hundred low-power transmitters. available for mobile cellular use was limited, efficient
 INTRODUCTION TO WIRELESS AND MOBILE NETWORKS / 153

use of the required frequencies was needed for mobile planning or frequency reuse. Frequency reuse was
cellular coverage. In modern cellular telephony, rural implemented by restructuring the mobile telephone
and urban regions are divided into areas according to system architecture into the cellular concept.
specific provisioning guidelines. Deployment The concept of frequency reuse is based on
parameters, such as amount of cell-splitting and cell assigning to each cell a group of radio channels used
sizes, are determined by engineers experienced in within a small geographic area. Cells are assigned a
cellular system architecture. group of channels that is completely different from
Provisioning for each region is planned according neighbouring cells. The coverage area of cells is called
to an engineering plan that includes cells, clusters, the footprint. This footprint is limited by a boundary so
frequency reuse, and handovers. that the same group of channels can be used in different
Cells: A cell is the basic geographic unit of a cellular cells that are far enough away from each other so that
system. The term cellular comes from the honeycomb their frequencies do not interfere.
shape of the areas into which a coverage region is
divided. Cells are base stations transmitting over small
geographic areas that are represented as hexagons. Each
cell size varies depending on the landscape. Because of
constraints imposed by natural terrain and man-made
structures, the true shape of cells is not a perfect
hexagon.
Clusters: A cluster is a group of cells. No channels
are reused within a cluster. Figure below, illustrates a
seven-cell cluster.

Cluster size
is expressed as n
In this cluster n = 7

Cells with the same number have the same set of

frequencies. Here, because the number of available
frequencies is 7, the frequency reuse factor is 1/7.
That is, each cell is using 1/7 of available cellular
channels.
Cell Splitting: Unfortunately, economic
considerations made the concept of creating full systems
with many small areas impractical. To overcome this
difficulty, system operators developed the idea of cell
splitting. As a service area becomes full of users, this
Frequency Reuse: Because only a small number approach is used to split a single area into smaller ones.
of radio channel frequencies were available for mobile In this way, urban centres can be split into as many areas
systems, engineers had to find a way to reuse radio as necessary to provide acceptable service levels in
channels to carry more than one conversation at a time. heavy-traffic regions, while larger, less expensive cells
The solution the industry adopted was called frequency can be used to cover remote rural regions.
154 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

The call continues as long as the user is talking, and the

Large cells
user does not notice the handoff at all.
for rural 4. North American Analog Cellular Systems:
areas Originally devised in the late 1970s to early 1980s,
analog systems have been revised somewhat since that
time and operate in the 800-MHz range. A group of
government, telco, and equipment manufacturers
worked together as a committee to develop a set of rules
(protocols) that govern how cellular subscriber units
(mobiles) communicate with the cellular system. System
development takes into consideration many different,
and often opposing, requirements for the system, and
often a compromise between conflicting requirements
Handoff: The final obstacle in the development
results. Cellular development involves the following
of the cellular network involved the problem created
basic topics:
when a mobile subscriber travelled from one cell to
frequency and channel assignments
another during a call. As adjacent areas do not use the
type of radio modulation
same radio channels, a call must either be dropped or
maximum power levels
transferred from one radio channel to another when a
modulation parameters
user crosses the line between adjacent cells. Because
messaging protocols
dropping the call is unacceptable, the process of
call-processing sequences
handoff was created. Handoff occurs when the mobile
The Advanced Mobile Phone Service (AMPS):
telephone network automatically transfers a call from
AMPS was released in 1983 using the 800-MHz to 900-
radio channel to radio channel as a mobile crosses
adjacent cells. MHz frequency band and the 30-kHz bandwidth for each
channel as a fully automated mobile telephone service.
It was the first standardized cellular service in the world
Handoff
and is currently the most widely used standard for
cellular communications. Designed for use in cities,
to Base
Station 2
AMPS later expanded to rural areas. It maximized the
cellular concept of frequency reuse by reducing radio
power output. The AMPS telephones (or handsets) have
the familiar telephone-style user interface and are
compatible with any AMPS base station. This makes
mobility between service providers (roaming) simpler
Cellular Switch: for subscribers. Limitations associated with AMPS
DMS-MTX include the following:
Public Switched low calling capacity
Telephone
Network (PSTN) limited spectrum
Trunk Routes
no room for spectrum growth
poor data communications
During a call, two parties are on one voice channel. minimal privacy
When the mobile unit moves out of the coverage area inadequate fraud protection
of a given cell site, the reception becomes weak. At this AMPS is used throughout the world and is
point, the cell site in use requests a handoff. The system particularly popular in the United States, South America,
switches the call to a stronger-frequency channel in a China, and Australia. AMPS uses frequency modulation
new site without interrupting the call or alerting the user. (FM) for radio transmission. In the United States,
 INTRODUCTION TO WIRELESS AND MOBILE NETWORKS / 155

transmissions from mobile to cell site use separate The Cell Site: The term cell site is used to refer to
frequencies from the base station to the mobile the physical location of radio equipment that provides
subscriber. coverage within a cell. A list of hardware located at a
Narrowband Analog Mobile Phone Service cell site includes power sources, interface equipment,
(NAMPS): Since analog cellular was developed, radio frequency transmitters and receivers, and antenna
systems have been implemented extensively throughout systems.
the world as first-generation cellular technology. In the Mobile Subscriber Units (MSUs): The mobile
second generation of analog cellular systems, NAMPS subscriber unit consists of a control unit and a transceiver
was designed to solve the problem of low calling that transmits and receives radio transmissions to and
capacity. NAMPS is now operational in 35 U.S. and from a cell site. The following three types of MSUs are
overseas markets, and NAMPS was introduced as an available:
interim solution to capacity problems. NAMPS is a U.S. the mobile telephone (typical transmit power
cellular radio system that combines existing voice is 4.0 watts)
processing with digital signaling, tripling the capacity the portable (typical transmit power is 0.6
of today’s AMPS systems. The NAMPS concept uses watts)
frequency division to get 3 channels in the AMPS 30- the transportable (typical transmit power is 1.6
kHz single channel bandwidth. NAMPS provides 3 watts)
users in an AMPS channel by dividing the 30-kHz The mobile telephone is installed in the trunk
AMPS bandwidth into 3 10-kHz channels. This
of a car, and the handset is installed in a
increases the possibility of interference because channel
convenient location to the driver. Portable and
bandwidth is reduced.
transportable telephones are hand-held and can
5. Cellular System Components: The cellular
be used anywhere. The use of portable and
system offers mobile and portable telephone stations
transportable telephones is limited to the
the same service provided fixed stations over
charge life of the internal battery.
conventional wired loops. It has the capacity to serve
6. Digital Systems: As demand for mobile
tens of thousands of subscribers in a major metropolitan
telephone service has increased, service providers found
area. The cellular communications system consists of
that basic engineering assumptions borrowed from
the following four major components that work together
to provide mobile service to subscribers: wireline (landline) networks did not hold true in mobile
public switched telephone network (PSTN) systems. While the average landline phone call lasts at
mobile telephone switching office (MTSO) least 10 minutes, mobile calls usually run 90 seconds.
cell site with antenna system Engineers who expected to assign 50 or more mobile
mobile subscriber unit (MSU) phones to the same radio channel found that by doing
PSTN so they increased the probability that a user would not
The PSTN is made up of local networks, the get dial tone this is known as call-blocking probability.
exchange area networks, and the long-haul network that As a consequence, the early systems quickly became
interconnect telephones and other communication saturated, and the quality of service decreased rapidly.
devices on a worldwide basis. The critical problem was capacity. The general
Mobile Telephone Switching Office (MTSO): characteristics of time division multiple access (TDMA),
The MTSO is the central office for mobile switching. It Global System for Mobile Communications (GSM),
houses the mobile switching centre (MSC), field personal communications service (PCS) 1900, and code
monitoring, and relay stations for switching calls from division multiple access (CDMA) promise to
cell sites to wireline central offices (PSTN). In analog significantly increase the efficiency of cellular telephone
cellular networks, the MSC controls the system systems to allow a greater number of simultaneous
operation. The MSC controls calls, tracks billing conversations. Figure below shows the components of
information, and locates cellular subscribers. a typical digital cellular system.
156 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

TDMA uses the same frequency band and channel

allocations as AMPS. Like NAMPS, TDMA provides
three to six time channels in the same bandwidth as a
single AMPS channel. Unlike NAMPS, digital systems
Microwave
Fibre Optic
have the means to compress the spectrum used to
transmit voice information by compressing idle time
Microwave and redundancy of normal speech. TDMA is the digital
Fibre Optic standard and has 30-kHz bandwidth. Using digital voice
encoders, TDMA is able to use up to six channels in the
same bandwidth where AMPS uses one channel.
The advantages of digital cellular technologies over Extended Time Division Multiple Access
analog cellular networks include increased capacity and (ETDMA): The ETDMA standard claims a capacity
security. Technology options such as TDMA and CDMA of fifteen times that of analog cellular systems. This
offer more channels in the same analog cellular capacity is achieved by compressing quiet time during
bandwidth and encrypted voice and data. Because of conversations. ETDMA divides the finite number of
the enormous amount of money that service providers cellular frequencies into more time slots than TDMA.
have invested in AMPS hardware and software, This allows the system to support more simultaneous
providers look for a migration from AMPS to digital cellular calls.
analog mobile phone service (DAMPS) by overlaying Fixed Wireless Access (FWA): FWA is a radio-
their existing networks with TDMA architectures. based local exchange service in which telephone service
Time Division Multiple Access (TDMA): North is provided by common carriers. It is primarily a rural
American digital cellular (NADC) is called DAMPS and application that is, it reduces the cost of conventional
TDMA. Because AMPS preceded digital cellular wireline. FWA extends telephone service to rural areas
systems, DAMPS uses the same setup protocols as by replacing a wireline local loop with radio
analog AMPS. TDMA has the following characteristics: communications. Other labels for wireless access
IS 54 standard specifies traffic on digital voice include fixed loop, fixed radio access, wireless
channels telephony, radio loop, fixed wireless, radio access, and
initial implementation triples the calling Ionica. FWA systems employ TDMA or CDMA access
capacity of AMPS systems technologies.
capacity improvements of 6 to 15 times that of
AMPS are possible
many blocks of spectrum in 800 MHz and 1900
MHz are used
all transmissions are digital
TDMA/FDMA application 7. 3 callers per
radio carrier (6 callers on half rate later),
providing 3 times the AMPS capacity
TDMA is one of several technologies used in
wireless communications. TDMA provides each call
with time slots so that several calls can occupy one
bandwidth. Each caller is assigned a specific time slot. Personal Communications Service (PCS): The
In some cellular systems, digital packets of information future of telecommunications includes PCS. PCS at
are sent during each time slot and reassembled by the 1900 MHz (PCS 1900) is the North American
receiving equipment into the original voice components. implementation of digital cellular system (DCS) 1800
 INTRODUCTION TO WIRELESS AND MOBILE NETWORKS / 157

(GSM). Trial networks were operational in the United of radio channels to be used within a small geographic
States by 1993, and in 1994, the Federal area called a cell. Base stations in adjacent cells are
Communications Commission (FCC) began spectrum assigned channel groups which contain completely
auctions. As of 1995, the FCC auctioned commercial different channels than neighboring cells. The base
licences. In the PCS frequency spectrum, the operator’s station antennas are designed to achieve the desired
authorized frequency block contains a definite number coverage within the particular cell. By limiting the
of channels. The frequency plan assigns specific coverage area to within the boundaries of a cell, the
channels to specific cells, following a reuse pattern that same group of channels may be used to cover different
restarts with each nth cell. The uplink and downlink cells that are separated from one another by distances
bands are paired mirror images. As with AMPS, a large enough to keep interference levels within tolerable
channel number implies one uplink and one downlink limits. The design process of selecting and allocating
frequency (e.g. Channel 512 = 1850.2-MHz uplink channel groups for all of the cellular base stations within
paired with 1930.2-MHz downlink). a system is called frequency reuse or frequency planning.
CODE DIVISION MULTIPLE ACCESS (CDMA) Figure given on next page illustrates the concept
CDMA is a digital air interface standard, claiming of cellular frequency reuse, where cells labelled with
8 to 15 times the capacity of analog. It employs a the same letter use the same group of channels. The
commercial adaptation of military, spread-spectrum, frequency reuse plan is overlaid upon a map to indicate
single-sideband technology. Based on spread spectrum where different frequency channels are used. The
theory, it is essentially the same as wireline service the hexagonal cell shape shown in Figure is conceptual
primary difference is that access to the local exchange and is a simplistic model of the radio coverage for each
carrier (LEC) is provided via wireless phone. Because base station, but it has been universally adopted since
users are isolated by code, they can share the same the hexagon permits easy and manageable analysis of a
carrier frequency, eliminating the frequency reuse cellular system. The actual radio coverage of a cell is
problem encountered in AMPS and DAMPS. Every known as the footprint and is determined from field
CDMA cell site can use the same 1.25-MHz band, so measurements or propagation prediction models.
with respect to clusters, n = 1. This greatly simplifies Although the real footprint is amorphous in nature, a
frequency planning in a fully CDMA environment. regular cell shape is needed for systematic system design
CDMA is an interference-limited system. Unlike and adaptation for future growth. While it might seem
AMPS/TDMA, CDMA has a soft capacity limit; natural to choose a circle to represent the coverage area
however, each user is a noise source on the shared of a base station, adjacent circles cannot be overlaid
channel and the noise contributed by users accumulates. upon a map without leaving gaps or creating overlapping
This creates a practical limit to how many users a system regions. Thus, when considering geometric shapes
will sustain. Mobiles that transmit excessive power which cover an entire region without overlap and with
increase interference to other mobiles. For CDMA, equal area, there are three sensible choices–a square,
precise power control of mobiles is critical in an equilateral triangle, and a hexagon. A cell must be
maximizing the system’s capacity and increasing battery designed to serve the weakest mobiles within the
life of the mobiles. The goal is to keep each mobile at footprint, and these are typically located at the edge of
the absolute minimum power level that is necessary to the cell. For a given distance between the centre of a
ensure acceptable service quality. Ideally, the power polygon and its farthest perimeter points, the hexagon
received at the base station from each mobile should be has the largest area of the three. Thus, by using the
the same (minimum signal to interference). hexagon geometry, the fewest number of cells can cover
CELLULAR SYSTEM DESIGN a geographic region, and the hexagon closely
FUNDAMENTALS approximates a circular radiation pattern which would
Cellular radio systems rely on an intelligent occur for an omnidirectional base 03_57_104_final.fm
allocation and reuse of channels throughout a coverage PMFrequency Reuse 59 station antenna and free space
region. Each cellular base station is allocated a group propagation. Of course, the actual cellular footprint is
158 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

determined by the contour in which a given transmitter the contour in which a given transmitter serves the
serves the mobiles successfully. mobiles successfully.
When using hexagons to model coverage areas, When using hexagons to model coverage areas,
base station transmitters are depicted as either being in base station transmitters are depicted as either being in
the centre of the cell (center-excited cells) or on three the centre of the cell (centre-excited cells) or on three
of the six cell vertices (edge-excited cells). Normally, of the six cell vertices (edge-excited cells). Normally,
omnidirectional antennas are used in center-excited cells omnidirectional antennas are used in centre-excited cells
and sectored directional antennas are used in corner- and sectored directional antennas are used in corner-
excited cells. Practical considerations usually do not excited cells. Practical considerations usually do not
allow base stations to be placed exactly as they appear allow base stations to be placed exactly as they appear
in the hexagonal layout. Most system designs permit a in the hexagonal layout. Most system designs permit a
base station to be positioned up to one-fourth the cell base station to be positioned up to one-fourth the cell
radius away from the ideal location. radius away from the ideal location.
To understand the frequency reuse concept, To understand the frequency reuse concept,
consider a cellular system which has a total of S duplex consider a cellular system which has a total of S duplex
channels available for use. If each cell is allocated a channels available for use. If each cell is allocated a
group of k channels (k < S), and if the S channels are group of k channels (k < S), and if the S channels are
divided among N cells into unique and disjoint channel divided among N cells into unique and disjoint channel
groups which each have the same number of channels, groups which each have the same number of channels,
the total number of available radio channels can be the total number of available radio channels can be
expressed as equation (i). expressed as
The N cells which collectively use the complete S = kN ...... (i)
set of available frequencies is called a cluster. If a cluster The N cells which collectively use the complete
is replicated M times within the system, the total number set of available frequencies is called a cluster. If a cluster
of duplex channels, C, can be used as a measure of is replicated M times within the system, the total number
capacity and is given by equation (ii). of duplex channels, C, can be used as a measure of
capacity and is given by
C = MkN = MS ...... (ii)
As seen from equation (ii), the capacity of a
cellular system is directly proportional to the number
of times a cluster is replicated in a fixed service area.
The factor N is called the cluster size and is typically
equal to 4, 7, or 12. If the cluster size N is reduced
while the cell size is kept constant, more clusters are
required to cover a given area, and hence, more
capacity (a larger value of C) is achieved. A large
cluster size indicates that the ratio between the cell
radius and the distance between co-channel cells is
small. Conversely, a small cluster size indicates that
co-channel cells are located much closer together. The
value for N is a function of how much interference a
mobile or base station can tolerate while maintaining
a sufficient quality of communications.
From a design viewpoint, the smallest possible
value of N is desirable in order to maximize capacity
Station antenna and free space propagation. Of
over a given coverage area (i.e. to maximize C in
course, the actual cellular footprint is determined by
 INTRODUCTION TO WIRELESS AND MOBILE NETWORKS / 159

equation (ii)). The frequency reuse factor of a cellular Dynamic channel assignment strategies require the MSC
system is given by 1/N, since each cell within a cluster to collect real-time data on channel occupancy, traffic
is only assigned 1/N of the total available channels in distribution, and radio signal strength indications (RSSI)
the system. of all channels on a continuous basis. This increases
Channel Assignment Strategies: For efficient the storage and computational load on the system but
utilization of the radio spectrum, a frequency reuse provides the advantage of increased channel utilization
scheme that is consistent with the objectives of and decreased probability of a blocked call.
increasing capacity and minimizing interference is Handoff Strategies: When a mobile moves into a
required. A variety of channel assignment strategies have different cell while a conversation is in progress, the
been developed to achieve these objectives. Channel MSC automatically transfers the call to a new channel
assignment strategies can be classified as either fixed belonging to the new base station. This handoff
or dynamic. The choice of channel assignment strategy operation not only involves identifying a new base
impacts the performance of the system, particularly as station, but also requires that the voice and control
to how calls are managed when a mobile user is handed signals be allocated to channels associated with the new
off from one cell to another. base station.
In a fixed channel assignment strategy, each cell is Processing handoffs is an important task in any
allocated a predetermined set of voice channels. Any cellular radio system. Many handoff strategies prioritize
call attempt within the cell can only be served by the handoff requests over call initiation requests when
unused channels in that particular cell. If all the channels allocating unused channels in a cell site. Handoffs must
in that cell are occupied, the call is blocked and the be performed successfully and as infrequently as
subscriber does not receive service. Several variations possible, and be imperceptible to the users. In order to
of the fixed assignment strategy exist. In one approach, meet these requirements, system designers must specify
an optimum signal level at which to initiate a handoff.
called the borrowing strategy, a cell is allowed to borrow
Once a particular signal level is specified as the
channels from a neighboring cell if all of its own
minimum usable signal for acceptable voice quality at
channels are already occupied. The mobile switching
the base station receiver (normally taken as between–
center (MSC) supervises such borrowing procedures and
90 dBm and -100 dBm), a slightly stronger signal level
ensures that the borrowing of a channel does not disrupt
is used as a threshold at which a handoff is made.
or interfere with any of the calls in progress in the donor
This margin, given by  = Pr handoff–Pr minimum
cell.
usable cannot be too large or too small. If  is too large,
In a dynamic channel assignment strategy, voice
unnecessary handoffs which burden the MSC may occur,
channels are not allocated to different cells permanently.
and if  is too small, there may be insufficient time to
Instead, each time a call request is made, the serving
complete a handoff before a call is lost due to weak
base station requests a channel from the MSC. The signal conditions. Therefore,  is chosen carefully to
switch then allocates a channel to the requested cell meet these conflicting requirements. Figure below
following an algorithm that takes into account the illustrates a handoff situation. Figure (a) demonstrates
likelihood of future blocking within the cell, the the case where a handoff is not made and the signal
frequency of use of the candidate channel, the reuse drops below the minimum acceptable level to keep the
distance of the channel, and other cost functions. channel active. This dropped call event can happen when
Accordingly, the MSC only allocates a given there is an excessive delay by the MSC in assigning a
frequency if that frequency is not presently in use in the handoff or when the threshold  is set too small for the
cell or any other cell which falls within the minimum handoff time in the system. Excessive delays may occur
restricted distance of frequency reuse to avoid co- during high traffic conditions due to computational
channel interference. Dynamic channel assignment loading at the MSC or due to the fact that no channels
reduce the likelihood of blocking, which increases the are available on any of the nearby base stations (thus
trunking capacity of the system, since all the available forcing the MSC to wait until a channel in a nearby cell
channels in a market are accessible to all of the cells. becomes free).
160 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Illustration of a handoff senior at Cell boundary

In deciding when to handoff, it is important to should be made quickly. Information about the vehicle
ensure that the drop in the measured signal level is not speed, which can be useful in handoff decisions, can
due to momentary fading and that the mobile is actually also be computed from the statistics of the received
moving away from the serving base station. In order to short-term fading signal at the base station.
ensure this, the base station monitors the signal level
for a certain period of time before a handoff is initiated.
This running average measurement of signal strength Q. 1. Explain the term ‘Airport’ in Wire-
should be optimized so that unnecessary handoffs are lessTransmission.
avoided, while ensuring that necessary handoffs are Ans. Airport is standard (802.11) and has a faster
completed before a call is terminated due to poor signal transfer rate. Technologically it is more advanced, more
level. The length of time needed to decide if a handoff rugged and smaller and capable of being used anywhere.
is necessary depends on the speed at which the vehicle It is 10 mb/sec, 50 metres, 10 active devices. Apparently,
is moving. If the slope of the short-term average received Wireless transmisson is 1mb/sec, 10 metres and 7 active
signal level in a given time interval is steep, the handoff devices.
 INTRODUCTION TO WIRELESS AND MOBILE NETWORKS / 161

Q. 2. How frequency hopping is used for security include Point-to-Point Protocol, TCP/IP, UDP, OBEX
in Wireless transmisson? Protocol, and WAP.
Ans. Wireless transmission technology uses Q. 5. What is Message-Oriented Middleware
Adaptive Frequency Hopping and capable to reduce (MOM)?
interference between wireless technologies with the help Ans. An infrastructure focuses on sending and
of 2.4 GHz spectrum. In order to take the advantage of receiving messages to increment interoperability,
the available frequency, Blue tooth’s AFH works within flexibility and flexibility of an application. MOM
the spectrum and is performed by the technology performs this by allowing an application to be distributed
over platforms of different kind. MOM reduces the
detecting other devices in the spectrum and avoiding
application development complexity which spans
the used frequencies. Efficient transmission within the
multiple operating systems and network protocols. This
spectrum at high degree of interference immunity is
process insulates the application developer from the
achieved by adapting hopping among 79 frequencies at
operating systems details and network interfaces.
1 MHz intervals. Various APIs across diverse platforms and networks are
Q. 3. Explain how data is transferred through provided by MOM.
Wireless transmission. Q. 6. What are TP-Lite and TP-Heavy
Ans. The PSI line wireless transmission interface Monitors?
converters are used making complex data connections Ans. TP-Lite Monitor: The integration of TP
with mobile devices with the utilization of flexible monitors functions in a database engines is called as
conduits, slip rings and other traditional cable solutions. TP-Lite monitor.
The universal PSI line interface makes it possible TP-Heavy Monitor: A TP monitor that supports
to transmit wireless transmission of the interfaces such the client/server architecture and allows PC for initiating
as RS-232, RS-422, RS-485 wire and USB. very complex multiserver transaction from the desktop.
Q. 4. Explain how Wireless transmisson devices Q. 7. What is Wireless transmisson?
integrate easily with TCP/IP network. Ans. Wireless transmission is a wireless technology
Ans. Wireless transmisson is an IEEE 802.11 standard used to exchange data over short distances.
standard and a replacement of wired network. It is inter The data is exchanged from fixed and mobile devices
operable and works as an Adopted protocol, VOIP, by creating Personal Area Network with security at high
ADVTP, RFCOMM and L2TP. Much telephony related level. Wireless transmisson was originally conceived
profiles are related to it as a carrier for DOT/AT as a wireless alternate to RS-232 cables. It resolves the
problems synchronization by connecting several
commands, and also being a transport layer for OBEX
devices.
over Wireless transmisson.
Q. 8. Wireless transmisson is called a cable
RFCOMM is being used by many Bluetooth
replacement technology. Explain.
applications and is wide spread for support and publicly
Ans. Wireless transmisson allows Personal Area
available API on most of the operating systems. Quick Networks without the cables. It provides connectivity
porting of applications that uses serial port for to many mobiles users at a time for sharing without
communication can be done by using RFCOMM. wires. Blue tooth chip is designed for replacing cables
Adopted protocols are defined by other by transmitting the information at a special frequency
organizations of standards-making and are incorporated from sender to receiver.
into the protocol stack of Bluetooth. These protocols
 162

Network Security
16
an overview of the requirements for network security.
The requirements of information security within an Next, we look at conventional encryption and its use in
organization have undergone two major changes in the providing privacy; this is followed by a discussion of
last several decades. Before the widespread use of data message authentication. We then look at the use of
processing equipment, the security of information felt public-key encryption and some of the mathematical
to be valuable to an organization was provided primarily background for them.
by physical and administrative means; an example of
the former is the use of rugged filing cabinets with a INTRODUCTION TO COMPUTER SECURITY
combination lock for storing sensitive documents; an The protection afforded to an automated
example of the latter is personnel screening procedures information system in order to attain the applicable
used during the hiring process. objectives of preserving the integrity, availability and
With the introduction of the computer, the need for confidentiality of information system resources
automated tools for protecting files and other (includes hardware, software, firmware, information/
information stored on the computer became evident; this data, and telecommunications).
is especially the case for a shared system, such as a time- The Basic Components
sharing system, and the need is even more acute for Computer security rests on confidentiality, integrity,
systems that can be accessed over a public telephone or and availability. The interpretations of these three
data network. The generic name for the collection of aspects vary, as do the contexts in which they arise. The
tools designed to protect data and to thwart hackers is interpretation of an aspect in a given environment is
computer security. Although this is an important topic, dictated by the needs of the individuals, customs, and
it is beyond the scope of this book and will be dealt laws of the particular organization.
with only briefly. 1. Confidentiality: Confidentiality is the
The second major change that affected security is concealment of information or resources. The need for
the introduction of distributed systems and the use of keeping information secret arises from the use of
networks and communications facilities for carrying data computers in sensitive fields such as government and
between terminal user and computer and between industry. For example, military and civilian institutions
computer and computer. Network security measures are in the government often restrict access to information
needed to protect data during their transmission, and to to those who need that information. The first formal
guarantee that data transmissions are authentic. work in computer security was motivated by the
The essential technology underlying virtually all military’s attempt to implement controls to enforce a
automated network and computer security applications “need to know” principle. This principle also applies to
is encryption. Two fundamental approaches are in use: industrial firms, which keep their proprietary designs
conventional encryption, also known as symmetric secure lest their competitors try to steal the designs. As
encryption, and public-key encryption, also known as a further example, all types of institutions keep personnel
asymmetric encryption. As we look at the various records secret.
approaches to network security, these two types of Access control mechanisms support confidentiality.
encryption will be explored. The chapter begins with One access control mechanism for preserving

Receiver
 NETWORK SECURITY / 163

confidentiality is cryptography, which scrambles data Integrity mechanisms fall into two classes:
to make it incomprehensible. A cryptographic key prevention mechanisms and detection mechanisms.
controls access to the unscrambled data, but then the Prevention mechanisms seek to maintain the
cryptographic key itself becomes another datum to be integrity of the data by blocking any unauthorized
protected. attempts to change the data or any attempts to change
Other system-dependent mechanisms can prevent the data in unauthorized ways. The distinction between
processes from illicitly accessing information. Unlike these two types of attempts is important. The former
enciphered data, however, data protected only by these occurs when a user tries to change data which she has
controls can be read when the controls fail or are no authority to change. The latter occurs when a user
bypassed. Then their advantage is offset by a authorized to make certain changes in the data tries to
corresponding disadvantage. They can protect the change the data in other ways. For example, suppose an
secrecy of data more completely than cryptography, but accounting system is on a computer. Someone breaks
if they fail or are evaded, the data becomes visible. into the system and tries to modify the accounting data.
Confidentiality also applies to the existence of data, Then an unauthorized user has tried to violate the
which is sometimes more revealing than the data itself. integrity of the accounting database. But if an accountant
The precise number of people who distrust a politician hired by the firm to maintain its books tries to embezzle
may be less important than knowing that such a poll money by sending it overseas and hiding the
was taken by the politician's staff. How a particular transactions, a user (the accountant) has tried to change
government agency harassed citizens in its country may data (the accounting data) in unauthorized ways (by
be less important than knowing that such harassment moving it to a Swiss bank account). Adequate
occurred. Access control mechanisms sometimes authentication and access controls will generally stop
conceal the mere existence of data, lest the existence the break-in from the outside, but preventing the second
itself reveal information that should be protected.
type of attempt requires very different controls.
Resource hiding is another important aspect of
Detection mechanisms do not try to prevent
confidentiality. Sites often wish to conceal their
violations of integrity; they simply report that the data’s
configuration as well as what systems they are using;
integrity is no longer trustworthy. Detection mechanisms
organizations may not wish others to know about
may analyze system events (user or system actions) to
specific equipment (because it could be used without
detect problems or (more commonly) may analyze the
authorization or in inappropriate ways), and a company
data itself to see if required or expected constraints still
renting time from a service provider may not want others
to know what resources it is using. Access control hold. The mechanisms may report the actual cause of
mechanisms provide these capabilities as well. the integrity violation (a specific part of a file was
All the mechanisms that enforce confidentiality altered), or they may simply report that the file is now
require supporting services from the system. The corrupt.
assumption is that the security services can rely on the Working with integrity is very different from
kernel, and other agents, to supply correct data. Thus, working with confidentiality. With confidentiality, the
assumptions and trust underlie confidentiality data is either compromised or it is not, but integrity
mechanisms. includes both the correctness and the trustworthiness of
2. Integrity: Integrity refers to the trustworthiness the data. The origin of the data (how and from whom it
of data or resources, and it is usually phrased in terms was obtained), how well the data was protected before
of preventing improper or unauthorized change. it arrived at the current machine, and how well the data
Integrity includes data integrity (the content of the is protected on the current machine all affect the integrity
information) and origin integrity (the source of the data, of the data. Thus, evaluating integrity is often very
often called authentication). The source of the difficult, because it relies on assumptions about the
information may bear on its accuracy and credibility source of the data and about trust in that source-two
and on the trust that people place in the information. underpinnings of security that are often overlooked.
This dichotomy illustrates the principle that the aspect 3. Availability: Availability refers to the ability to
of integrity known as credibility is central to the proper use the information or resource desired. Availability is
functioning of a system. We will return to this issue when an important aspect of reliability as well as of system
discussing malicious logic.
164 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

design because an unavailable system is at least as bad intermediaries. Messages may also be kept in
as no system at all. The aspect of availability that is repositories, such as message queues or databases. Some
relevant to security is that someone may deliberately of the data within the messages is considered to be
arrange to deny access to data or to a service by making sensitive in nature. There is a risk that an attacker can
it unavailable. System designs usually assume a gain access to sensitive data, either by eavesdropping
statistical model to analyze expected patterns of use, on the network or accessing a repository.
and mechanisms ensure availability when that statistical Use encryption to protect sensitive data that is
model holds. Someone may be able to manipulate use contained in a message. Unencrypted data, which is
(or parameters that control use, such as network traffic) known as plaintext, is converted to encrypted data, which
so that the assumptions of the statistical model are no is known as ciphertext. Data is encrypted with an
longer valid. This means that the mechanisms for algorithm and a cryptographic key. Ciphertext is then
keeping the resource or data available are working in converted back to plaintext at its destination.
an environment for which they were not designed. As a 4. Authorization: Also known as access control,
result, they will often fail. this form of security decides who is allowed where. File
Attempts to block availability, called denial of permissions are a good example of access controls.
service attacks, can be the most difficult to detect, Usually stored in resource-level ACLs (access control
because the analyst must determine if the unusual access lists), these are simply lists of authenticated users (or
patterns are attributable to deliberate manipulation of groups of users) who are allowed to access or are barred
resources or of environment. Complicating this from accessing a given resource. This is usually the most
determination is the nature of statistical models. Even complex aspect of network security since it requires the
if the model accurately describes the environment, secure, centralized storage and access of these ACLs.
atypical events simply contribute to the nature of the Authorization can occur only once a user is
statistics. A deliberate attempt to make a resource authenticated, as authorization systems rely on lists of
unavailable may simply look like, or be, an atypical authenticated users.
event. In some environments, it may not even appear AUTHENTICATION AND PRIVACY
atypical. This is simple verification of a user's identity.
SECURITY SERVICES Always based on some form of trust, authentication
Following are the network security services: relies on something that the user has, which can be
1. Authentication: The process of identifying an compared to a known constant (the trusted value). This
individual, usually based on a username and password. can occur either in an interaction with the user (a user
In security systems, authentication is distinct from ID and password are entered), or can range up to
authorization , which is the process of giving individuals complex biometrics systems like fingerprint
access to system objects based on their identity. identification, face recognition or retinal scans.
Authentication merely ensures that the individual is who Authentication also can occur by proxy, such as a stored
he or she claims to be, but says nothing about the access authentication token (kept in a workstation's memory
rights of the individual. while the user remains logged in, o r on a token such as
2. Integrity: In information security, integrity a smartcard).
means that data cannot be modified undetectably. Cryptography is the heart of security. If, we need
Means refers to the validity of data. Data integrity to create privacy, we need to encrypt our message at the
can be compromised in a number of ways: sender site and decrypt it at the receiver site. The science
Human errors when data is entered of cryptography is very complex. A cryptographic expert
Errors that occur when data is transmitted from needs to be knowledgeable in areas such as mathematics,
one computer to another electronics and programming. The word cryptography
Software bugs or viruses in Greek means “secret writing”. Nowadays it refers to
Hardware malfunctions, such as disk crashes the science and art of transforming messages to make
Natural disasters, such as fires and floods them secure and immune to attacks. The figure below
3. Confidentiality: Data passes between a client shows the components of cryptography:
and a Web service, sometimes through one or more The original message before being transmitted, is
 NETWORK SECURITY / 165

Sender Receiver

Plaintext Plaintext
Encryption Network Dencryption
ciphertext ciphertext

called plaintext. After the message is transformed, it is operates on. To encrypt a message, we need encryption
called cipher-text. An encryption algorithm transforms algorithm, an encryption key and plaintext. To decrypt
the plaintext to cipher-text; a decryption algorithm the message, we need a decrypt key and a cipher text.
transforms the cipher-text back to plaintext. The sender BLOCK AND STREAM CIPHERS
uses an encryption algorithm and the receiver uses a Traditional ciphers used a character or symbol as
decryption algorithm. the unit of encryption/decryption. Modern ciphers, on
We use encryption/decryption algorithms as the other hand, use a block of bits as the unit of
encryption/decryption. In the figure given below, shows
ciphers. The term cipher is also used to refer to different
the concept of the block ciphers, the plaintext and cipher
categories of algorithms or operations in cryptography.
text are block of bits.
A key is a number (value) that the cipher, as an algorithm,

S-Boxes: Substitutions are performed by S-boxes, too large an effect. For these applications, a fourth
as shown in Fig. below. In this example a 3-bit plaintext option, stream cipher mode, exists. It works by
is entered and a 3-bit ciphertext is output. The 3-bit input encrypting an initialization vector, using a key to get
selects one of the eight lines exiting from the first stage an output block. The output block is then encrypted,
and sets it to 1; all the other lines are 0. using the key to get a second output block. This block
is then encrypted to get a third block, and so on. The
(arbitrarily large) sequence of output blocks, called
the keystream, is treated like a one-time pad and XORed
with the plaintext to get the ciphertext, as shown in fig
(a). Note that the IV is used only on the first step.
After that, the output is encrypted. Also note that the
keystream is independent of the data, so it can be
computed in advance, if need be, and is completely
Stream Cipher: Applications exist in which having insensitive to transmission errors. Decryption is
a 1-bit transmission error mess up 64 bits of plaintext is shown in fig. (b).
166 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

Decryption occurs by generating the same encryption algorithm parameterized by Alice’s public
keystream at the receiving side. Since the keystream key. Similarly, the (secret) decryption algorithm
depends only on the IV and the key, it is not affected by parameterized by Alice’s private key is DA. Bob does
transmission errors in the ciphertext. Thus, a 1-bit error the same thing, publicizing EB but keeping DB secret.
in the transmitted ciphertext generates only a 1-bit error We can solve the problem of establishing a secure
in the decrypted plaintext. channel between Alice and Bob, who have never had
PUBLIC AND PRIVATE KEY CRYPTOGRAPHY any previous contact. Both Alice’s encryption key, EA,
Distributing the keys has always been the weakest and Bob’s encryption key, EB, are assumed to be in
link in most cryptosystems. Keys had to be protected publicly readable files. Now Alice takes her first
from theft, but they also had to be distributed, so they message, P, computes EB(P), and sends it to Bob. Bob
could not just be locked up in a bank vault. In 1976, then decrypts it by applying his secret key DB [i.e. he
two researchers at Stanford University, Diffie and computes DB(EB(P)) = P]. No one else can read the
Hellman (1976), proposed a radically new kind of encrypted message, EB(P), because the encryption
cryptosystem, one in which the encryption and system is assumed strong and because it is too difficult
decryption keys were different, and the decryption key to derive DB from the publicly known EB. To send a
could not feasibly be derived from the encryption key. reply, R, Bob transmits EA(R). Alice and Bob can now
In their proposal, the (keyed) encryption algorithm, E, communicate securely.
and the (keyed) decryption algorithm, D, had to meet A note on terminology is perhaps useful here.
three requirements. These requirements can be stated Public-key cryptography requires each user to have two
simply as follows: keys: a public key, used by the entire world for
1. D(E(P)) = P. encrypting messages to be sent to that user, and a private
2. It is exceedingly difficult to deduce D from E. key, which the user needs for decrypting messages. We
3. E cannot be broken by a chosen plaintext will consistently refer to these keys as the public and
attack. private keys, respectively, and distinguish them from
The first requirement says that if we apply D to an the secret keys used for conventional symmetric-key
encrypted message, E(P), we get the original plaintext cryptography.
message, P, back. Without this property, the legitimate INTRODUCTION TO RSA
receiver could not decrypt the ciphertext. The second RSA (Rivest-Shamir-Adleman Algorithm)
requirement speaks for itself. The third requirement is
Due to the potential advantages of public-key
needed because, as we shall see in a moment, intruders
cryptography, many researchers are hard at work, and
may experiment with the algorithm to their hearts'
some algorithms have already been published. It is
content. Under these conditions, there is no reason that
known by the initials of the three discoverers (Rivest,
the encryption key cannot be made public.
Shamir, Adleman): RSA. Much practical security is
The method works like this: A person, say, Alice,
wanting to receive secret messages, first devices two based on it. Its major disadvantage is that it requires
algorithms meeting the above requirements. The keys of at least 1024 bits for good security (versus 128
encryption algorithm and Alice’s key are then made bits for symmetric-key algorithms), which makes it quite
public, hence the name public-key cryptography. Alice slow.
might put her public key on her home page on the Web, The RSA method is based on some principles from
for example. We will use the notation EA to mean the number theory:
 NETWORK SECURITY / 167

1. Choose two large primes, p and q (typically 1. Alice and Bob have to agree on two large
1024 bits). numbers, n and g, where n is a prime, (n – 1)/2
2. Compute n = p × q and z = (p – 1) × (q – 1). is also a prime and certain conditions apply to
3. Choose a number relatively prime to z and g. These numbers may be public, so either one
call it d. of them can just pick n and g and tell the other
4. Find e such that e × d = 1 mod z. openly. Now Alice picks a large (say, 512-bit)
Encryption: number, x, and keeps it secret. Similarly, Bob
1. Divide the plaintext (regarded as a bit string) picks a large secret number, y.
into blocks, so that each plaintext message, P, 2. Alice initiates the key exchange protocol by
falls in the interval 0 ? P < n. Do that by sending Bob a message containing (n, g, gx
grouping the plaintext into blocks of k bits,
mod n), as shown in fig. Bob responds by
where k is the largest integer for which 2k < n
is true. sending Alice a message containing gy mod n.
2. To encrypt a message, P, compute C = Pe (mod
n).
3. To decrypt C, compute P = Cd (mod n).
4. It can be proven that for all P in the specified
range, the encryption and decryption functions
are inverses. To perform the encryption, you
need e and n. To perform the decryption, you
need d and n. Therefore, the public key consists
of the pair (e, n), and the private key consists
of (d, n).
Difficulties associated, according to Rivest and 3. Now Alice raises the number Bob sent her to
colleagues, factoring a 500-digit number requires 1025 the xth power modulo n to get (gy mod n)x
years using brute force. In both cases, they assume the mod n. Bob performs a similar operation to
best known algorithm and a computer with a 1-µsec get (gx mod n)y mod n. By the laws of modular
instruction time. arithmetic, both calculations yield gxy mod n.
For example, how the RSA algorithm works is given Alice and Bob suddenly share a secret key,
in fig, we have chosen p = 3 and q = 11, giving n = 33 gxy mod n.
and z = 20. A suitable value for d is d = 7, since 7 and For example, an intruder, Trudy, has seen both
20 have no common factors. With these choices, e can messages. She knows g and n from message 1. If she
be found by solving the equation 7e = 1 (mod 20), which could compute x and y, she could figure out the secret
yields e = 3. The ciphertext, C, for a plaintext message, key. The trouble is, given only gx mod n, she cannot
P, is given by C = P3 (mod 33). The ciphertext is find x. No practical algorithm for computing discrete
decrypted by the receiver by making use of the rule P = logarithms modulo a very large prime number is known.
C7 (mod 33). The figure shows the encryption of the For example, we use the values of n = 47 and g = 3.
plaintext “SUZANNE” as an example. Alice picks x = 8 and Bob picks y = 10. Both of these
Diffie–Hellman are kept secret. Alice’s message to Bob is (47, 3, 28)
The protocol that allows strangers to establish a because 38 mod 47 is 28. Bob’s message to Alice is
shared secret key is called the Diffie-Hellman key
exchange (Diffie and Hellman, 1976) and works as (17). Alice computes 178 mod 47, which is 4. Bob
follows:
168 / NEERAJ : FUNDAMENTALS OF COMPUTER NETWORK

computes 2810 mod 47, which is 4. Alice and Bob have a function of the right input and the key for
independently determined that the secret key is now 4. this stage, Ki. All the complexity lies in this
Trudy has to solve the equation 3x mod 47 = 28, which function. for this stage, Ki. All the complexity
can be done by exhaustive search for small numbers lies in this function.
like this, but not when all the numbers are hundreds of The function consists of four steps, carried out
bits long. All currently-known algorithms simply take in sequence. First, a 48-bit number, E, is
too long, even on massively parallel supercomputers. constructed by expanding the 32-bit Ri - 1
The disadvantage of the Diffie–Hellman algorithm, according to a fixed transposition and
there is a problem: when Bob gets the triple (47, 3, 28), duplication rule. Second, E and Ki are XORed
how does he know it is from Alice and not from Trudy? together. This output is then partitioned into
There is no way he can know. eight groups of 6 bits each, each of which is
DES AND MD5 AT THE BEGINERS LEVEL fed into a different S-box. Each of the 64
DES - Data Encryption Standard possible inputs to an S-box is mapped onto a
This cipher, DES (Data Encryption Standard), was 4-bit output. Finally, these 8 x 4 bits are passed
widely adopted by the industry for use in security through a P-box.
products was developed by IBM in 1970. Plaintext is In each of the 16 iterations, a different key is
encrypted in blocks of 64 bits, giving 64 bits of used. Before the algorithm starts, a 56-bit
ciphertext. The algorithm, which is parameterized by a transposition is applied to the key. Just before
56-bit key, has 19 distinct stages. The first stage is a each iteration, the key is partitioned into two
key-independent transposition on the 64-bit plaintext. 28-bit units, each of which is rotated left by a
The last stage is the exact inverse of this transposition. number of bits dependent on the iteration
The stage prior to the last one exchanges the leftmost number. Ki is derived from this rotated key by
32 bits with the rightmost 32 bits. The remaining 16 applying yet another 56-bit transposition to it.
stages are functionally identical but are parameterized A different 48-bit subset of the 56 bits is
by different functions of the key. The algorithm has been extracted and permuted on each round.
designed to allow decryption to be done with the same A technique that is sometimes used to make
key as encryption, a property needed in any symmetric- DES stronger is called whitening. It consists
key algorithm. The steps are just run in the reverse order. of XORing a random 64-bit key with each
Each stage takes two 32-bit inputs and plaintext block before feeding it into DES and
produces two 32-bit outputs. The left output then XORing a second 64-bit key with the
is simply a copy of the right input. The right resulting ciphertext before transmitting it.
output is the bitwise XOR of the left input and Whitening can easily be removed by running
 NETWORK SECURITY / 169

the reverse operations (if the receiver has the Standard 8732, is shown in fig. below. Here two keys
two whitening keys). and three stages are used. In the first stage, the
Triple-DES (3 DES): In 1979, IBM realized that plaintext is encrypted using DES in the usual way
the DES key length was too short and devised a way with K1. In the second stage, DES is run in decryption
to effectively increase it, using triple encryption mode, using K2 as the key. Finally, another DES
developed by (Tuchman, 1979). The method chosen, encryption is done with K1.
which has since been incorporated in International

This design immediately gives rise to two questions.

First, why are only two keys used, instead of three? Q. 1. What is data encryption?
Second, why is EDE (Encrypt Decrypt Encrypt) used, Ans. Data encryption ensures data safety and very
instead of EEE (Encrypt Encrypt Encrypt)? The reason important for confidential or critical data. It protect data
that two keys are used is that even the most paranoid from being read, altered or forged while transmission.
cryptographers believe that 112 bits is adequate for Q. 2. What is the public key encryption?
routine commercial applications for the time being. Ans. Public key encryption use public and private
The reason for encrypting, decrypting, and then key for encryption and decryption. In this mechanism,
encrypting again is backward compatibility with existing public key is used to encrypt messages and only the
single-key DES systems. Both the encryption and corresponding private key can be used to decrypt them.
decryption functions are mappings between sets of 64- To encrypt a message, a sender has to know recipient's
bit numbers. public key.
MD5: A variety of message digest functions have Q. 3. Define digital signatures.
been proposed. The most widely used ones are MDS Ans. Digital signature is an attachment to an
and SHA-1. MD5 is the fifth is a series of message electronic message used for security purpose. It is used
digests designed by Rouald Rivest. It operates by to verify the authenticity of the sender.
mangling bits in a sufficiently complicated way that Q. 4. What is IGMP protocol?
every output bit is affected by every input bit. It starts Ans. Internet Group Management Protocol, allows
out by padding he message to a length of 448 bits. Then internet hosts to multicast. i.e. to send messages to a
the original length of the message is appended as a 64 group of computers. There may be a group of internet
bit interger to give a total input whose length is a multiple hosts interested to multicast. IGMP allows router to
of 5/Z bits. The last precomputation step is initailizing determine which host groups have members on a given
a 128 bit buffer to a fixed value. network segment. It helps to establish group
MD5 has been around for over a decade new, and memberships. It is commonly used for streamlining
many people have attacked i. Some vulnerabities have videos and gaming. The protocol can be implemented
been found, but certain internal steps prevent it from both as a host side and router side. The host side is
being broken. However, if the remaiming borners within responsible to notify its membership in a group. The
MD5 fall, it may eventually fail. Never-the-less at the notification is made to a local router. This local router
time of this writing it was still standing. (router side) in turn sends out queries.