DATA COMMUNICATION AND NETWORKING

1. Introduction
2. Data Communication
3. Multiplexing
4. Components of Computer Network
5. Local Area Network (LAN)
6. Uses of a Network
7. Topology
8. Networking Cables
9. OSI Layout and IEEE
10. Popular Types of LAN
11. Inter Networking
12. Public Networks
13. Client/Server vs. Peer to Peer Networking
14. Network Operating Systems
15. Network Management
16. LAN in Business Environments
17. Summary
18. Keywords
19. Review Questions
20. Further Readings

INTRODUCTION

 Today world is often referred to as global village. All thanks to today’s

revolutionary technology.
 Because of today’s technology, we can communicate to any portion ion any corner
of the world from anywhere in the world.
 And the credit for all this goes to the communication revolution that has happened
in this century, especially in the last thirty years.
 Advances in the communication technology, have made possible nearly everything
we dreamt of. It is basically communication that makes you take advantage of
massive mainframe processing power if you turn your PC into a terminal of a
mainframe.
 It is communication that lets you chat with different people across the globe
through an on-line services. It is again communication that allows you to access
huge servers of information on virtually any subject via networks of computers.
And that is why this chapter is dedicated to this fascinating combination of
communication, computer and networking.
DATA COMMUNICATION
 Flow of information for purpose of efficient management and business process
control requires effective use of computer systems and networking technologies.
Even for a simple important exercise of printing any information from a computer
by an attached printer to a computer, there is a requirement for error free data
transmission from the computer to the printer.
 The whole concept of data communication is based upon principles of sending
data, checking and confirming its receipt following error correction technique to
ensure accurate data transmission.
 This communication is possible with the approach known as protocols.
 The data flows between two devices as sequential bits (0 and 1) across a
transmission medium say copper wire.
 These bits are grouped to form a character. A character may be a number, an
alphabet or a symbol. As a standard, generally 8 bits form a byte or a character.
Each bit (0 or 1) is represented as a voltage level on a transmission medium.
 As a standard, bit 0 represents a positive voltage and bit 1 represents a negative
voltage.
 These signalling conventions are followed by most of the communicating devices.
The most common interface standards are RS232c or V5, X.21 etc. In RS232c
signalling standard, voltage level between +5V to + 15V represents bit 0 and
voltage level between -5 to -15V represents bit 1. During an idle state while the
data is not being transmitted, only the electric current flows through a transmission
medium (wire). At the instance of data transmission the current in the wireline is
intercepted in a particular way to superimpose data that is representable as the
voltage levels to traverse towards receiving device.

Asynchronous Transmission
 In asynchronous mode of data transmission each character (byte) is transmitted
one at a time.
 Each character is preceded by a start bit.
 The purpose of start bit is to inform the receiving device that a character
transmission has begun.
 The transmitted character ends with a stop bit.
 The stop bit suggests that a character has been transmitted.
 Each bit represents a voltage level for a specific length of time to enable the
receiving device to recognize and record the incoming signal accurately.
 This specific length of time determines the rate of data transfer that may be
measurable in bits per second.
 In asynchronous mode of transmission there is no indication of the speed at which
the character is being transmitted. Therefore, there is no perceptible method for the
 receiving device to sample the incoming signal precisely to recover the transmitted
data. For this lack of synchronization of transmission this mode of transmission is
called asynchronous transmission.
 This anamoly of non-synchronization is overcome by providing internal clocks at
each end, transmitting as well as receiving. Both internal clocks work at the same
rate such that on the receipt of start bit the receiver device starts its internal clock
to instruct receiver's interface to sample the incoming signal.

Synchronous Transmission
 In synchronous mode of data transmission a block of characters is transmitted one
at a time. Each block of characters is preceded with a few synchronization (SYN)
characters.
 These SYN characters contain the timing information which is used by the
receiving device clock onto the sending device's signal. Since, in this mode of data
transfer the block contains the timing information using SYN characters that
allows automatic synchronization, this process is known as synchronous
transmission.
 The size of blocks of characters may vary therefore there exist the ASCII control
characters termed as STX (start of text) and ETX (end of text) that are used as
delimiters.
 The synchronous mode of data transmission is advantageous when there exists
large volume of data to be transmitted at a higher data transfer rate.
 The efficiency of synchronous transmission varies according to the block size and
dependent upon the protocol. If the block size is too small the overheads of
protocol is more due to additional SYN characters and delimiters in header
although the probability of error in transmission reduces. In case the block size is
too large the general efficiency for large data transfer may improve in terms of
data transfer rate but the probability of error in transmission increases.
 Therefore, for a reliable error free data transfer over a long distance requires that a
maximum block size be fixed. Frequently used block size standard in most of the
protocol is 512 bytes. Generally the level of efficiency in the synchronous mode of
transmission is higher than in the asynchronous mode of transmission.
Error Management in Data Transmission
 In synchronous transmission, the most common approach of error detection in data
transmission is CRC (Cyclic Redundancy Check) or check sum. This is adopted as
a CCITT standard.
 CCITT (Consultive Committee for International Telephony and Telegraphy) is an
international committee of telecom standards based in France.
 A checksum that is a remainder of data is divided by a fixed polynomial or a bit
pattern is appended to the data bits before transmission. At the receiving device
the data is verified against the checksum. For recovery of erroneous data packets
in general and an error free data transfer in particular, generally it is implicit for
the receiving device to send a copy of data back to the sending device as a means
of acknowledging correct or incorrect receipt.
 In the case of a satellite link connection a technique known as FEC (Forward Error
Correction) is followed primarily for memory access and also for data
transmission and recovery. This approach deters response back to the sending
device by incorporating some additional information embedded in the data. The
receiving device makes the necessary correction automatically by the
identification of these codes in the embedded data. These codes are also referred to
as Hamming codes.

Notes: Issues of Interoperability

 The structure of the unit of data sent should be recognized at the receiving end;
  The representation of data (ASCII/EBCDIC) should be the same on both sides of
data communications;
 There should be a method of introducing checkpoints for recovery during large
data transfer in the event of failure;
 There should be adequate security measures;
 There should be end to end protocol in both communicating hosts to recover from
errors in a similar fashion;
 Both communicating hosts should use similar addressing techniques. Approaches
for Data Transmission

Approaches for Data Transmission

 In a network of communicating devices there are two popular approaches for data
transmission.
1. virtual circuit
2. datagram service
 First approach is to open a connection across the network and let the data pass
through that connection.
 At this connection, the routing information is established and route numbers are
assigned to blocks of data. This connection is also referred as a virtual circuit.
 Another approach is a connectionless approach.
 In this approach all packets carry full address of its destination and sequence
identity. Each packet takes routes independently without a pre-established route
and reassemble and re-sequence at the receiving device. This approach is also
referred to as a datagram service.

TYPES OF DATA TRANSMISSION

Data transmission can be divided into the following two types:

 Serial transmission
 Parallel transmission
 1. Serial transmission is the transmission of data one bit behind another along a single wire.

2. Parallel transmission is the transmission of data along a set of several wires (usually a multiple of
eight wires) where bits move simultaneously as a bit front.

 Parallel transmission is faster than serial transmission.

 Parallel transmission is less reliable than serial transmission because of the fine tolerances in the
transmission as the bits can become muddled up.

Examples:
1. The Data transmission between the primary memory and the hard drive of the computer is
parallel because the processor requires the data to be downloaded as quickly as possible.
2. The data transmission between the computer and the printer is parallel because the computer
needs to send a big chunk of data to the printer’s buffer quickly so that the processor will be free
to attend to other tasks of the computer.

Modes of data transmission

Data transmission can be divided into the following three modes:
1. Duplex mode
2. Half duplex mode
3. Simplex mode

Duplex mode of transmission

 Means passing data in both directions at the same time between two devices.
 The data transmission between the primary memory and the hard drive of the computer is half
duplex.

Half duplex mode of transmission

 Means passing data in both directions between two devices but only one direction at a time.
 The data transmission between the computer and the printer is half duplex because the computer
transmit data files to the printer for printing and the printer passes messages to the computer such
as error messages.
Simplex mode of transmission
 Means passing data only in one direction between two devices.

Bit rate
 Number of bits that can be sent in one second is known as bit rate.
 The unit used to measure the bit rate is Baud.
 1 Baud = 1 bit per second
What is the relationship between the bit rate and the way that the data in the files are used?
1. When files containing only text data are transmitted in a network, they can be sent at a slower bit
rate because the text data are not time sensitive.
2. When files containing pictures, graphics and animations are transmitted in a network, they have
to be sent at a higher bit rate because those data are time sensitive.
3. When files other than the ones containing only text data is transmitted in a network, it is
important to limit the amount of data that is being transmitted. This will increase the speed of data
transmission and therefore the time it takes to download the file at the destination computer will
decrease. Speeding up the transmission of a file by reducing the amount of data is known as
compression.
Why bit rate is an important part of communication?
This is because it enables:
1. Standardization
2. One layer of the protocol to be altered without altering the entire protocol
3. Manufacturers to design hardware and software for a particular layer

Why bit rate is an important part of any protocol?

 Because if one device sends data at a particular bit rate and the other device reads what it receives
at a different rate, the message would never be received correctly.

Error checking and correcting

When data is transmitted in a network the bit stream can get affected due to various reasons,
which makes the data at the receiving computer to contain errors. The methods that can be used
for checking transmission errors are:
1. Echoing back
2. Check sum
3. Parity
ECHOING BACK
 The receiving computer sending back the data to the sending computer is known as echoing back.

How echoing back is used to detect transmission errors?

 The data echoed back is checked against the data that was sent in the first place by the sending
computer.
 If they match there has been no error in transmission. If they do not match there has been an error
in transmission and the data will be sent again.
Disadvantages of echoing back
 Takes a long time to transmit data as the data has to be sent twice
 Can be implemented only in duplex or half duplex modes

CHECK SUM
 The sum of all the bytes in a block of data is called the check sum.
How check sum is used to detect transmission errors?
 The check sum is calculated ignoring the carry out of the byte and is mentioned with the block of
data that is being transmitted.
 The check sum is then calculated again when it is received at the receiving computer
 The check sum calculated is then compared with the value mentioned in the block of data.
 If the two values match there has been no error in transmission. If they do not match there has
been an error in transmission.

PARITY

 The sum of all the 1-bits in a byte of data is called the parity. Parity can be either odd or even.
 The parity of each byte of data must be equal to the parity defined for that computer.
 If the parity of a byte of particular piece of data is different from the parity of the computer, then
the parity bit of the byte is changed so that the parity of the data will match the parity of the
computer.

How parity check is used to detect data transmission errors?

 The parity of each byte of data that reaches the receiving computer is checked against the parity
of the sending computer.
 If the two parities match there has been no error in transmission. If they do not match there has
been an error in transmission.

CIRCUIT SWITCHING
 Circuit switching has a fixed path through the network between the sending computer and the
destination computer is opened and held for the entire time duration of data transmission.
The advantages of circuit switching?
 The message arrives at the receiving computer without having its data to be reordered.

PACKET SWITCHING
 The message is split into a number of equal-sized packets.
 Each packet is labeled with the destination address and the packet number.
 Packet switching transmits individual data packets into the network to find their way to the
destination computer independently of each other.
 At each node the label is checked and the packet is redirected
 At the destination computer the packets are reordered and reassembled according to their packet
numbers.
The advantages of packet switching?

 Large part of the network is not tied up for long periods of time.
 Message is difficult to be hacked

The disadvantages of packet switching?

 Transmits at the speed of the slowest packet
 Packets must be reordered and reassembled at the destination

Protocol
 A protocol is a set of rules which control the data transmission in a network.

Features (or rules) of protocols?

 Type of data transmission or whether the data transmission is serial or parallel. Errors would
occur if one device has a serial connection and the other device expects a parallel connection.
 Mode of data transmission or whether the data transmission is full duplex, half duplex or simplex
 Bit rate – since both devices must talk and listen at the same number of bits per second.
Otherwise bits may be missed or counted twice
 Error checking – whether the parity is odd or even, whether echoing back is used, whether check
sums are used
 RS232c interface is typically designed to connect any digital equipment to a data
communication equipment. A modem is a typical DCE.

Lines of Communications
 There are basically two popular lines for communications.
i. A public telephone is the ordinary voice grade line. This line is routed
through many relay systems and electromechanical systems. This line
of communication is also referred to as PSTN (Public Switched
Telephone Network).
 ii. A leased line of communications is generally not subjected to the
level of switching as in PSTN. This line of communications
accommodates wider range of frequencies than the voice grade. A
leased line of communications consists of bidirectional
simultaneously operating two telephone circuits.

MULTIPLEXING
 Multiplexing is a technique of channelling multiple data inputs into one
connection.
 The approaches used to implement multiplexing is generally based on
division of communication channel into time slots or frequency bands.
 This technique was developed to overcome deficiency of old mainframe
machine which had direct terminal connections access for various
peripherals and other computers connected to it.
 Mainframe machine served as host to these terminals. This led to poor
utilization of the system because majority of connected systems remain
inactive for major period of time whereas some of the systems were
saturated with requirement of data access or processing at a specific time.
Moreover the feasibility to provide port for each terminal as requirements
grew was vague.
COMPONENTS OF COMPUTER NETWORK
 The typical hardware in a network of computers consists of the following
components:
a. computers,
b. peripherals such as printers, scanners, plotters and modems, cables,
network interface cards,
c. hubs and switches,
d. repeaters, bridges, routers, gateways.

LOCAL AREA NETWORK (LAN)

 A LAN is a network of computers that are interconnected in a particular
fashion and contained within a limited area say a building.
 A typical LAN has two components:
a) A host computer that is called ‘File server’. Large networks may have more
than one Fileservers.
b) Workstations or the terminals of a LAN are interconnected and connected to
the Fileserver.
 Workstations in a LAN are personal computers that have their own
processing power.
 The significant feature of a LAN is that its workstations are intelligent
terminals with a capability of distributed processing. This is unlike in a
mainserver or miniframe centralized server based environment where the
connected terminals are dumb without processing power of their own.

The prominent reasons which led to development and popularity of LANs are:
1. Sharing of expensive resources such as disk, files, printers, plotters, modems
and software was needed.
2. Diskless nodes (workstations) of a LAN provided security and virus
protection by not allowing to download important data from server or
uploading unwanted software.
3. Distributed processing for inter-user communication within an organization
and information exchange was the most suitable model for MIS
development.
4. Data management in a LAN environment is more efficient than the
centralized processing systems. Since, no users directly request to server for
data without resorting to file transfer on a stored media like floppy or hard
disk.
 5. LANs are more flexible and modular. This implies that LANs can be
expanded easily by adding of workstation with the support of networking
software. If more disk storage is needed, another hard disk can be installed
or other file server can be connected in a LAN. More than one LAN can be
bridged to share data and communicate.
6. Most LANs have fault tolerant features such that in case of power failures
LANs are capable of disk mirroring and disk duplexing to safeguard data.

USES OF A NETWORK
 The network provides an intelligent switching capability between the
devices for purpose of sharing. For instance an expensive laser printer in
network can be share be by many microcomputers in a network.
 The network enable creation of a large buffer where data accumulates from
various computing devices. In case of print sharing the buffer activates as a
queue on a special device known as print server.
 In a typical office environment the facilities of terminal access to central
computing machine such as mainframe, access to external and remote
devices, access to database or messaging or e-mail provisions necessitate the
networking of devices.
 The environments like point of sales in retail outlets, central heating
thermostats and time clocks, automatic telling machines to banks main
computers, process control devices in a manufacturing environment are not
feasible without networking.

TOPOLOGY
 The structure or layout of the cabling in connecting devices together to form
a network is called topology of the network.
 The advantage of a structured topology is that it is easier to add or delete a
node in the network.
Bus
 The Bus consists of a single communication channel.
 Each connected device is attached to the media at an interface point.
 Each connected device has a unique hardware address. Data transfer takes
place using these hardware addresses. The bus will have more complex
arrangements with several interconnected segments.
Advantages
 Simple layout
 Ease in connectivity
 Locating cable faults is easy
 Ideal for one to many data transmissions since all connected devices see the
traffic on the cable. The signal reaches all stations regardless of their
position on the bus.
Bottlenecks
 There is no automatic acknowledgement of receipt by virtue of the topology.
 The signals stop on reaching the end of the cable and do not return by
default to the sender. Therefore, the higher level protocols exists which can
ensure that data is successfully exchanged.

Ring
 Unlike bus topology, in a ring configuration the data transmission is
unidirectional.
 Data transmissions are received by each station’s interface as the data passes
through the interface connection. Like bus topology, each interface only
copies the data from the network. Thereafter data communicates to the
device connected to the network on basis of the packet’s destination address.
Each network interface connection has its own hardware address for
identification.
 Typical ring topology is generally used in the large network Fibre
Distributed Data Interface (FDDI).
 In ring topology, each interface station allows data transmission to pass
through the interface with only a few bits delay. For instance, consider
circulation of a data packet (say 128 bytes), the start of data packet may well
have circulated around the ring and arrived back at the originating station
before the end of the packet has left for circulation.
 There is always only one circulating data packet on the ring.
Advantages
 No routing problems since everyone on the ring receives the data. The
addressed node makes copy of the data for circulation
 Every generated packet eventually returns to the node therefore the
acknowledgment of successful data transmission is trivial
 Larger networks may be constructed from a number of interconnected rings.

Bottlenecks
 In case one of the nodes fails network comes to a halt.

Star
 Generally, it includes a central switching system or hub. In the figure, if
station D wishes to talk to station B, it does so via the hub. In case a new
device is required to be added it interfaces with another cable attached to the
central point. And if the hub is physically close to the connected device then
the individual cable rum would not be problematic.
 Star topology

Advantages
 Better overall reliability against cable faults provided the central point which
is switching data to the destination is robust
 Only a single node will be unavailable in the event of cable failure
 LANs implemented on the star topology are sometimes based on existing
telephone wiring in the building and do not normally have high data transfer
rates.
Bottlenecks
 In case the switch or hub fails the entire network is halted.

NETWORKING CABLES