BASIC ELEMENTS OF A COMMUNICATION SYSTEM

Communication is the process of transferring messages from one point to another. As

shown in Figure 17.1, the three basic elements of any communication process are:
1. A sender (source), which creates the message to be transmitted.
2. A medium, which carries the message.
3. A receiver (sink), which receives the message.
For example, when you speak to your friend on the telephone, you are the sender. The
telephone line through which your voice is transmitted is the medium and your friend is the
receiver. This is a simple example of voice communication. The same concept holds good for
data communication also. Data communication is the function of transporting data from one
point to another. In this case, the sender and the receiver are normally machines, in particular,
computer devices (computers, terminals, peripheral devices like line printers, plotters, disks,
etc.) and the transmission medium may he telephone lines, microwave links, satellite links, etc.
However, the messages that are transmitted are data, not voice conversations. Thus, the
electronic systems that transfer data from one point to another are called data communication
systems. Unlike computers that process and rearrange data, data communication systems
transmit data from one point to another without any change.

DATA TRANSMISSION MODES

There are three ways, or modes, of transmitting data from one point to another. As
shown in Figure 17.2, these are simplex, half-duplex, and full-duplex.

Simplex
If transmission is simplex, communication can take place in only one direction. Devices
connected to such a circuit are either a send-only or a receive-only device. For example, a data
collection terminal on a factory floor (send only) or a line printer (receive only). At first thought,
that might appear adequate for many types of applications in which flow of information is
unidirectional. However, in almost all data processing applications, communication in both
directions is required. Even for a "one-way" flow of information from a terminal to a computer,
the system will be designed to allow the computer to signal the terminal that data has been
received. Without this capability, the remote user might enter data and never know that it was
not received by the computer (due to some problem somewhere). Hence, simplex circuits are
seldom used because a return path is generally needed to send acknowledgement control, or
error signals.

Half-Duplex
A half-duplex system can transmit data in both directions, but only in one direction at a
time. Thus, a half-duplex line can alternately send and receive data. It requires two wires. This is
the most common type of transmission for voice communication because only one person is
supposed to speak at a time. It is also used to connect a terminal with a computer. The terminal
might transmit data and then the computer responds with an acknowledgement.

Full-Duplex
In a half-duplex system, the line must be "turned-around" each time the direction is
reversed. This involves a special switching circuit and requires a small amount of time
(approximately 150 milliseconds). With high-speed capabilities of the computer, this turn-around
time is unacceptable in many instances. Also, some applications require simultaneous
transmission in both directions. In such cases, a full-duplex system is used that allows
information to flow simultaneously in both directions on the transmission path. Use of a fullduplex line improves efficiency as the line turn-around time required in a half-duplex
arrangement is eliminated. It requires four wires.

Multiplexers
There are many applications in which several terminals are connected to a computer. If
each terminal is at 300 bits per second over a communication line (channel) that can operate at
9600 bits per second, then we see a very inefficient operation. It has been found that the
capacity of a channel exceeds that required for a single signal. A channel is an expensive
resource. Hence for its optimal utilization, the channel can be shared in such a way so as to
simultaneously transmit multiple signals over it. The method of dividing a physical channel into
multiple logical channels, so that a number of independent signals may be simultaneously
transmitted on it is known as multiplexing. The electronic device that performs this task is known
as a multiplexer.
A multiplexer takes several data communication lines or signals and converts them into
one data communication line or signal at the sending location. For example, as shown in Figure
17.11, there may be 4 terminals connect to a multiplexer, The multiplexer takes the signals from
the 4 terminals and coverts them into one large signal which can be transmitted over one
communication line. Then, at the receiving location, a multiplexer takes one large signal and
breaks it into the original 4 signals. Without multiplexers, you would have to have 4 separate
communication lines.

Thus, with multiplexing it is possible for a single transmission medium to concurrently

transmit data between several transmitters and receivers. There are two basic methods of
multiplexing channels. They are frequency-division multiplexing and time-division multiplexing.
Frequency-Division Multiplexing (FDM) - The bandwidth or range of a medium exceeds that
of any given signal. This fact is utilized for frequency-division multiplexing. In FDM, the available
bandwidth of a physical medium is split up (divided) into several smaller, disjoint logical
bandwidths. Each of the component bandwidths is used as a separate communication line
(channel). Figure 17.12 illustrates the process of FDM.
The best example of FDM is the way we receive various stations in a radio. Each radio
station is assigned a frequency range within a bandwidth of radio frequencies. Several radio
stations may be transmitting speech signals simultaneously over the physical channel, which is
"ether" in this case. A radio receiver antenna receives signals transmitted by all the stations.
Finally, the tuning dial in the radio is used to isolate the speech signal of the station tuned. In
FDM, the signals to be transmitted must be analog signals. Thus, digital signals must be
converted to analog form if they are to use FDM.

2. Time-Division Multiplexing (TDM) - The bit rate of transmission medium always exceeds
the required rate of the digital signal. This fact is utilized for time-division multiplexing. In TDM,
the total time available in the channel is divided between several users and each user of the
channel is allotted a time slice (a small time interval) during which he/she may transmit a
message. That is, multiple data streams belonging to different users are interleaved into one
data stream at the sending end of the channel. At the receiving end, a demultiplexer is used to
reassemble individual chunks of messages sent by the users into full messages. Figure 17.13
shows multiplexing of three different signals using TDM. The same idea may be extended for
multiplexing several signals.
Although TDM may be used to multiplex digital or analog signals, its usage is more
appropriate for digital data transmission. It is also very much suitable for communication
between computers because such communications occur in short fast bursts. Besides this, TDM
is generally more efficient as more sub channels can be derived. It is up to the network
designers to allocate time slots to different channels. It is common to have 32 low-speed
terminals connected to one high-speed line.
Whether or not to use multiplexing usually depends upon economics. The cost of highspeed modems and multiplexers is very high compared to the cost of low-speed modems.
However, if line costs are high due to long distances, then multiplexing is cost effective. One
serious disadvantage with multiplexing relates to a transmission line failure. If the line goes out,
everything is dead. With individual lines only one terminal is likely to be lost.
DATA TRANSMISSION MEDIA
Twisted Pair Wire Consists of two copper wires, enclosed separately in a plastic insulation,
twisted around each other. Also called Unshielded Twisted-Pair (UTP) because it is unshielded
from outside interference.
Used in short distance communication (upto 1 km)
data transmission speed 9,600 bits/sec (<100 mts) & 1,200 bits/sec (>100mts)
inexpensive & easy to install
use is limited and have comparatively high error rates
UTP cables are commonly used in local telephone communication and short distance (up
to about 1 km) digital data transmission. They are normally used to connect terminals to the
main computer if they are placed at a short distance from the main computer. Data transmission
speed of up to 9600 bits per second (or 9600 baud) can be achieved if the distance is not more
than 100 meters. However, for longer distance data transmission, local telephone lines are
used. In this case, typical speed of digital signal transmission is 1200 bits per second.
UTP cables are an inexpensive medium of data transmission. They are easy to install and use.
However, their use is limited because they easily pick up noise signals, which results in high
error rates when the line length extends beyond 100 meters.

Coaxial Cable
Coaxial cables are groups of specially wrapped and insulated wire lines that are able to
transmit data at high rates. As shown in Figure 17.4, they consist of a central copper wire
surrounded by a PVC insulation over which a sleeve of copper mesh is placed. The metal
sleeve is again shielded by an outer shield of thick PVC material. The signal is transmitted by
the inner copper wire and is electrically shielded by the outer metal sleeve.
Coaxial cables offer much higher bandwidths than UTP cables and are capable of transmitting
digital signals at rates of 10 mega bits per second. They are extensively used in long distance
telephone lines arid as cables for cable TV. They are also used by telephone companies to
transmit data. In many cases, several coaxial cables are packaged into a very large cable that
can handle over 40,000 telephone calls simultaneously. Furthermore, coaxial cables have much
higher noise immunity and can offer cleaner and crisper data transmission without distortion or
loss of signal.

Microwave System
Another popular transmission media is microwave. This is a popular way of transmitting
data since it does not incur the expense of laying cables. Microwave systems use very high
frequency radio signals to transmit data through space. However, at microwave frequencies, the
electromagnetic waves cannot bend or pass obstacles, like hills. The sender and the receiver of
a microwave system, which are mounted on very high towers, should be in a line-of-sight. This
may not be possible for very long distance transmission. Moreover, the signals become weaker
after travelling a certain instance and therefore, use repeaters at intervals of about 25 to 30 kms.

Microwave systems permit data transmission rates of about 16 Giga (1 Giga = 10 9) bits
per second. Furthermore, it can carry literally thousands of voice channels at the same time.
The link can support 250,000 voice channels. The initial installation cost of microwave links
being very high, they are mostly used to link big cities with heavy telephone traffic between
them.

Communications Satellite
The main problem with microwave communication is that the curvature of the earth,
mountains, and other structures often block the line-of-sight. Due to this reason/several repeater
stations are normally required for long distance transmission, which increases the cost of data
transmission between two points. This problem is overcome by using satellites, which are
relatively newer, and more promising data transmission media.

A communication satellite is basically a microwave relay station placed in outer space.

These satellites are launched either by rockets or by space shuttles and are precisely positioned
36,000 kms above the equator with an orbit speed that exactly matches the earth's rotation
speed. Since a satellite is positioned in a geosynchronous orbit, it is stationary relative to earth
and always stays over the same point on the ground. This allows a ground station to aim its
antenna at a fixed point in the sky. Each satellite can receive and retransmit signals to slightly
less than half of the earth's surface. Therefore, at least three satellites are needed in
geosynchronous orbit to provide worldwide data transmission service. Hundreds of satellites are
now in orbit to handle international and domestic data, voice, and video communications needs.
The INSAT series of Indian satellites are positioned in such a way that they are accessible from
any place in India.

Optical Fibres
Optical fibres are hair-thin threads of glass or plastic that can serve as a data
transmission medium as copper wires or coaxial cables. The basic difference is that optical
fibres transmit light signals instead of electrical signals. Because light travels much faster than
electricity, optical fibres can transmit data at much higher speed than copper wires or coaxial
cables, and also with no significant loss of intensity over very long distances.
Optical fibres are made of glass, plastic or silica. Plastic fibres are least efficient, but tend to be
cheaper and more rugged. Glass or silica fibres are much smaller, and their lower attenuation
makes them more suited for very high capacity channels.
Physically, a fibre-optic cable consists of three concentric layers - the inner core, a
cladding around it, and the outer protective coating. The inner core, which has a diameter of 8 to
200 micrometers, consists of a bunch of optical fibres. The cladding around it is made of plastic
or glass and has a refractive index less than that of the core. The characteristics of light
propagation depend primarily on the fibre size, its construction, the refractive index profile, and
the nature of the light source. The outer protective coating is made up of plastic.
The main components of an optical fibre communication system are shown in Figure
17.7. Towards its source side is a converter that converts electrical signals into light waves. The
converter uses either a light-emitting diode (LED) or a laser diode to convert electric signals into
light signals. These light waves are then transmitted over the optical fibre to the receiver's end.
At the receiver's end, another converter is placed that detects the light waves and converts them
back to electrical signals. It uses photoelectric diodes for this purpose. These electric signals are
then amplified using an amplifier and sent to the receiver.

Network

A logical extension of a data communication system.

Two or more computers are linked together using a particular medium of communication.
Data and information are sent, received or shared by the connected nodes/computers.
The speed of the connection depends on the medium used and the distance between the
nodes.
Differentiated into
a) Local Area Network (LAN)
b) Wide Area Network (WAN)
c) Metropolitan Area Network (MAN)
Key characteristics often used to differentiate between the different types of networks
a) Geographic distribution
b) Data rate
c) Error rate
d) Communication link
e) Ownership
f) Communication cost

THE OSI MODEL

The initial computer networks had their own set of standards and conventions that were quite
hardware oriented. Each manufacturer used to develop their own communication protocols for their
networks. For example, IBM launched SNA (Systems Network Architecture) in 1974. Similarly, DEC
(Digital Equipment Corporation) launched its network in 1980 for use on the DEC range of computers.
The architecture of DECNET (Digital Equipment Corporation Network) is known as DNA (Digital Network
Architecture). Like SNA, it is a private network architecture oriented only to DEC hardware. Other
telecommunication protocols are BNA (Burroughs Network Architecture) by Burroughs Incorporation,
DSN (Distributed Systems Network) by Hewlett Packard, PRIMENET (Prime Computers Network) by
Prime Incorporation, etc. Thus the data communication protocols of one network were not compatible
with any other network. Moreover, standards of the same network architecture also kept changing from
time to time. For example, earlier versions of SNA were not compatible with its subsequent versions.
Such incompatibilities started creating bottleneck in the efficient and proper utilization of network
resources.
This problem was recognized by the International Standards Organization (ISO), which
established a subcommittee to develop an international standard on network architectures. The result
was the Open System Interconnection (OSI) model, which is a framework for defining standards for
linking heterogeneous computers in a packet switched network. Thus the standardized OSI protocols
made it possible for any two heterogeneous computer systems, located anywhere in the world, to easily
communicate with each other.
The OSI model is also designed in a highly structured way. Its architecture is shown in Figure
17.24. It is a seven-layer architecture in which a separate set of protocols is defined for each layer. Thus
each layer has an independent function and deals with one or more specific aspects of the
communication. The roles of the seven layers are briefly described below.

In actual implementation, of the seven layers, the first three layers are likely to be in hardware,
the next two layers in the operating system, the presentation layer in library subroutines in the user's
address space, and the application layer in the user's program.

PHYSICAL LAYER
The physical layer is responsible for transmitting raw bit streams between two nodes. That is, it
may convert the sequence of binary digits into electric signals, light signals, or electromagnetic signals
depending on whether the two nodes are on a cable circuit, fiber-optic circuit, or microwave/radio circuit,

respectively. Electrical details such as how many volts to use for 0 and 1, how many bits can be sent per
second, and whether transmission can take place only in one direction or in both directions
simultaneously are also decided by the physical layer protocols. In addition, the physical layer protocols
also deal with the mechanical details such as the size and shape of the connecting plugs, the number of
pins in the plugs, and the function of each pin. In short, the physical layer protocols deal with the
mechanical, electrical, procedural, and functional characteristics of transmission of raw bit streams
between two sites. RS232-C is a popular physical layer standard for serial communication lines.

DATA-LINK LAYER
The physical layer simply transmits the data from the sender's node to the receiver's node as raw
bits. It is the responsibility of the data-link layer to detect and correct any errors in the transmitted data.
Since the physical layer is only concerned with a raw bit stream, the data-link layer partitions it into
frames so that error detection and correction can be performed independently for each frame. The datalink layer also performs flow control of frames between two sites to ensure that a sender does not
overwhelm a receiver by sending frames at a rate faster than the receiver can process. Therefore, the
error control and flow control mechanisms of a network form the data-link layer protocols in the OSI
model. Notice that the data-link layer and physical layer protocols establish an error-free communication
of raw hits between two nodes.

NETWORK LAYER
The network layer is responsible for setting up a logical path between two nodes for
communication to take place. It encapsulates frames into packets that can be transmitted from one node
to another using a high-level addressing and routing scheme. That is, routing is the primary job of the
network layer and the routing algorithm forms the main part of the network layer protocols of the network.
Two popular network layer protocols are the X.25 Protocol and the Internet Protocol (called IP). The X.25
is a connection-oriented protocol that is based on the concept of establishing a virtual circuit between the
sender and receiver before the actual communication starts between them. In this protocol, a request for
connection is first sent to the destination, which can either be accepted or rejected. If the connection is
accepted, the requesting party is given a connection identifier to use in subsequent requests. During the
connection establishment phase, a route between the two parties is also decided that is used for the
transmission of subsequent traffic.
On the other hand, IP is a connectionless protocol in which no connection is established between
the sender and receiver before sending a message. Therefore, each packet of the message is
transmitted independently and may take a different route. IP is part of the DoD (U.S. Department of
Defence) protocol suite.
Notice that the functions performed at the network layer are primarily required in WANs. In a
single LAN, the network layer is largely redundant because packets can be transmitted directly from any
node on the network to any other node. Therefore the network layer, if present, has little work to do.

TRANSPORT LAYER
The job of the transport layer is to provide node-to-node communication and to hide all the details
of the communication subnet from the session layer by providing a network-independent transport
service. Using this service, all the details of the communication subnet are sealed and one subnet can be
replaced with another without disturbing the layers above the transport layer.
In particular, the transport layer accepts message of arbitrary length from the session layer,
segments them into packets, submits them to the network layer for transmission, and finally reassembles
the packets at the destination. Some packets may be lost on the way from the sender to the receiver,
and depending on the routing algorithms used in the network layer, packets may arrive at the destination
in a sequence that is different from the order in which they are sent. The transport layer protocols include
mechanisms for handling lost and out-of-sequence packets. For this, the transport layer records a
sequence number in each packet and uses the sequence numbers for detecting lost packets and for
ensuring that messages are reconstructed in the correct sequence.
The two most popular transport layer protocols are the Transport Control Protocol (TCP) and the
User Datagram Protocol (UDP). TCP is a connection-oriented transport protocol. It uses end-to-end
mechanisms to ensure reliable, ordered delivery of data over a logical connection. These goals are
basically achieved by using packet sequence numbers and positive acknowledgements with timeout and
retransmission.
The UDP is a connectionless transport protocol. It is an unreliable protocol because, when it is
used, message packets can be lost, duplicated, or arrive out of order. Therefore, only those applications
that do not need reliable communication should use UDP.

SESSION LAYER
The purpose of the session layer is to provide the means by which presentation entities can
organize and synchronize their dialog and manage their data exchange. It allows the two parties to
authenticate each other before establishing a dialog session between them. It also specifies dialog type one way, two way alternate, or two way simultaneous - and initiates a dialog session if the message is a
connection request message. The other services of the session layer include quarantine service, dialog
control, and priority management. The quarantine service buffers a group of messages on the receiving
side until the session layer on the sending side explicitly releases them. This is useful in database
applications where a transaction (consisting of a group of messages) needs to be an atomic unit. The
dialog control is useful for dialog sessions in which the user primitives used for sending and receiving

messages are of the non-blocking type. In this case, the user may have multiple requests outstanding on
the same session, and replies may come back in an order different from that in which the requests were
sent. The dialog' control reorders replies according to the order of requests. The priority management
service is useful for giving priority to important and time-bound messages over normal, less-important
messages. The session layer is not required for connectionless communication.

PRESENTATION LAYER
The purpose of this layer is to represent message information to communicating application layer
entities in a way that preserves meaning while resolving syntax differences. For this, the presentation
layer may perform one or more of die following types of transformations on message data;
1. A message usually contains structured information that may include any of the data types used in
programming languages - integers, characters, arrays, records, and so on, including user-defined
data types. Translation is therefore required where language systems or application programs in
the source and destination computers use different representations for these data types.
2. Data format conversions are also needed to transfer data between computers when the hardware
of the sending and receiving computers uses different data representations. In this case, the
presentation layer software in the sending computer transforms message data from the formats
used in its own computer to a set of standard network representations called external Data
Representation (XDR) before transmission.
The presentation layer software in the receiving
computer transforms the message data from the network representations to the formats used in its
own computer.
3. For applications dealing with confidential or secret data, the presentation layer software in the
sending computer encrypts message data before passing it to the session layer. On the receiver
side, the encrypted message data is decrypted by the presentation layer before being passed on to
the application layer.
4.
In a similar manner, when message data is large in volume (such as multimedia data) or with
networks that are slow or heavily loaded, message data may be compressed and decompressed
by the presentation layer software in the sending and receiving computers, respectively.

APPLICATION LAYER
The application layer provides services that direct!}' support the end users of the network.
Obviously, the functionality implemented at this layer of the architecture is application-specific. Since
each application has different communication needs, no fixed or standard set of application layer
protocols can meet the needs of all applications. Therefore, the application layer is basically a collection
of miscellaneous protocols for various commonly used applications such as electronic mail, file transfer,
remote login, remote job entry, and schemas for distributed databases. Some popular application layer
protocols are X.400 (Electronic Mail Protocol), X.500 (Directory Server Protocol), FTP (File Transfer
Protocol), and rlogin (Remote Login Protocol).