BEE4102 TELEPHONE SYSTEMS CHAPTER 3

CHAPTER III: DATA NETWORKS

A. INTRODUCTION
Data networks are classified according to their geographical coverage:

- Wide area networks (WANs)

- Metropolitan area networks (MANs)
- Local area networks (LANs)

Intercity, inter-country, and intercontinental networks are known as 'WANs'. Based on the
communication infrastructure used, they may be classified as terrestrial data networks (TDNs) or
satellite-based data networks (SBDNs). In TDNs, data communication is organized using cables,
fiber optic lines, or radio links. A geostationary or geosynchronous satellite is used for
communication in SBDNs.

A metropolitan area network interconnects computers within a metropolitan city. Community

antenna television (CATV) cables, twisted pair wires or shielded lines, optical fibers, radio links, or
line-of-sight (LOS) optical communication links provide the communication medium for MANs.
The broadband capability of CATV cables permits carrying voice, data, and video simultaneously.
Thus, MANs are usually multimedia networks. Local area networks are confined to a single building
or a group of buildings generally belonging to the same organization. Optical fibers, twisted pair,
and coaxial cables are used as the communication media for LANs. Fiber optic networks (FONs) are
suitable for both LANs and MANs. Synchronous optical networks (SONET) are designed to operate
at high speeds. LANs, MANs, and WANs are generally interconnected in a hierarchical manner to
form a global network as shown in the figure. LANs are often connected directly to WANs,
particularly in places where MANs are not installed or have not developed well. Apart from the
different geographical coverage, the range of data rates supported on these networks also differ
widely. The figure summarizes the typical data rates and geographical coverage for these networks.

Figure 3.1: Data network hierarchy

Page 1 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

B. DATA TRANSMISSION IN PSTNs

Public switched telephone networks (PSTNs) and electronic PABXs (Private Automated Branch
Exchanges) are designed primarily to carry analog voice signals. However, they can be utilized for
data transmission by incorporating suitable interfaces. As depicted in Figure 10.2, Local Area
Networks (LANs) can be constructed around PABXs, and Metropolitan Area Networks (MANs)
around PSTNs. In these scenarios, data rates are typically limited to a maximum of 64 kbps.
Terrestrial data networks and Integrated Services Digital Networks (ISDNs), however, support data
rates of 1.544 or 2.048 Mbps. A modulator translates the data pulses into voice-band analog signals
at the transmitting end. At the receiving end, the analog signals are demodulated to recover the digital
information. A combined modulator-demodulator unit is commonly referred to as a modem.

Figure 3.2: Geographical coverage and speeds of data networks

Initially, modems were utilized to connect terminals, located in remote places, to a central computer.
Later, computer-to-computer communication was established using modems and PSTNs. A data
communication scheme employing modems and a PSTN is illustrated in Figure 10.3. The digital
interface of a modem is linked to the computer, while the analog interface connects to the telephone
network. In addition to data exchange, the digital interface enables control signals to be exchanged
between the modem and the computer.

Page 2 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

Figure 3.3: Data communication using PSTN.

In a PSTN, a connection may be established using a dial-up facility or a dedicated non-switched

leased line. A 2-wire or 4-wire leased line may be employed. Modems typically feature automatic
dialing and answer facilities, which are useful when working with a dial-up line. These modems are
driven by communication software packages that operate on the terminal or the personal computer
of the user. The transmission may be half-duplex or full-duplex, depending on the configuration.

Data Rates in PSTNs

A voice channel in a PSTN typically has a nominal bandwidth of 3.1 kHz. According to Nyquist's
theorem, which applies to noiseless channels and states:

𝑅 = 2𝐻 log 2 𝑉 bps

Where:

- 𝑅 = maximum data rate

- 𝐻 = bandwidth of the channel
- 𝑉 = number of discrete levels in the signal

For a 3 kHz channel and a binary signal, the maximum data rate would theoretically be 6000 bps if
the channel were ideal. However, in a practical channel, the maximum rate would be lower.

Page 3 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

By increasing the number of levels used to represent the signal, the bit rate can be increased
arbitrarily in a noiseless channel. However, it's important to note that the actual number of signal
transitions is still limited to the binary level limit. The effective bit rate increases with more than two
signal levels because each signal level can represent a group of two or more bits. The maximum rate
of signal transitions that can be supported by a channel is known as the baud rate or symbol rate.

In a channel where noise is present, there is an absolute maximum limit for the bit rate. This limit
arises because the difference between two adjacent signal levels becomes comparable to the noise
level when the number of signal levels is increased. Claude Shannon extended Nyquist's work to the
case of noisy channels affected by random or thermal noise. Shannon's result states:

Figure 3.4: Baud rates and bit rates.

Modems
Amplitude, frequency, and phase modulation are all utilized in modem design. In amplitude
modulation, zeros and ones are represented by two different voltage levels. A signal waveform 𝑠(𝑡),
referred to as the baseband signal, is generated from the digital data. This signal is then multiplied
by a sinusoidal carrier, for example, cos(2𝜋𝑓0 𝑡), to produce a modulated signal.

Page 4 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

The high-frequency component at (2𝜋𝑓0 ) is then filtered out, leaving the demodulated signal 𝑠(𝑡)
from which the original digital signal may be retrieved. It is not necessary that only two voltage
levels are used when generating 𝑠(𝑡). A new digital message may be chosen at every symbol interval,
which is mapped to one of a set of discrete voltage levels. There is a one-to-one correspondence
between the discrete message set and the voltage set. The voltage levels are then used to modulate
the carrier. When more than two voltage levels are used, the modulation is known as M-ary. In an
M-ary scheme, a message sample contains ‘k’ bits, such that M = 2k. For example, in a 4-ary scheme,
each sample contains four bits of information.

While the message sample rate is the baud rate, the bit rate is four times the baud rate in this example.
This technique of varying the amplitude of the sinusoidal carrier using the voltage levels of the
baseband signal is known as amplitude shift keying (ASK). If the carrier is sinusoidal and the
baseband voltage level is used to vary the frequency or phase of the carrier, the modulation is known
as frequency shift keying (FSK) or phase shift keying (PSK), respectively. In FSK, two or more
different tones are used to represent zeros and ones or a group of message bits. Binary FSK modems
operating at 1200 bps on dial-up lines are widely used for personal computer communications.
Separate half-duplex channels are used for each direction. FSK has the advantage of a constant
amplitude signal, which makes it robust against the nonlinearities of power-limiting devices in the
channel.

C. SWITCHING TECHNIQUES FOR DATA TRANSMISSION

Telephone networks, originally designed to carry voice traffic, present significant disparities in
accommodating data traffic. Voice traffic typically exhibits continuous transmission (aside from
silent intervals in normal speech), while data traffic is inherently bursty. When a user interacts with
a computer terminal, there are distinct states: thinking, keying, or waiting. During the thinking and
waiting states, no data transmission occurs, whereas during the keying state, the user inputs data. To
minimize waiting time, user input must be promptly transmitted to the system after keying. Similarly,
once the machine computes a response, it must be swiftly transmitted back to the user. Consequently,
data traffic often requires high bandwidth for short durations, including transmissions between
computers.

In contrast, voice traffic requires low bandwidth (typically 3.4 kHz) for extended durations. Usually,
transmission lines remain idle for 85-95% of the holding time during data transmission and remain
active for a similar period during a telephone conversation (pauses in speech are considered active
transmission periods). Voice traffic typically operates in half-duplex mode, while data traffic may
operate in half or full-duplex. Additionally, acceptable error and loss rates differ significantly; data
transmission demands minimal errors or losses, while a small amount of speech loss may not be
noticeable.

Page 5 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

Furthermore, voice traffic typically occurs in real-time, while data traffic may or may not occur in
real-time. Interactive computer use necessitates real-time or near-real-time responses, whereas file
transfers between computers do not require real-time transmission. The table below summarizes the
disparities between voice and data traffic.

Voice Traffic Data Traffic

Continuous Bursty
Low bandwidth for long duration High bandwidth for short duration
Typical line utilization 85-95% Typical line utilization 5 – 15%
Half-duplex Half or full duplex
Real time Nonreal time or near real time
Loss acceptable Loss unacceptable
Error tolerable Error unacceptable

Circuit Switching
In circuit switching, an electrical path is established between the source and the destination before
any data transfer takes place. This electrical path can be realized through physical wires, coaxial
cables, radio, or satellite links. Once established, the path remains dedicated to the communicating
pair for the entire duration of the transmission, regardless of whether data is actually being transferred
or not. No other potential user can utilize the path, even if it is idle. The connection is released only
when specifically signaled by either of the communicating entities. Data transmission using a Public
Switched Telephone Network (PSTN) connection is a typical example of circuit-switched data
transfer. The figure illustrates the principle of circuit switching.

When the host computer H1 wants to transfer data to the host computer H5, a connection request is
made to the switching node N1, which, in turn, selects a suitable neighboring node through which
the desired connection may be established, for example, node N5. Node N5 then selects a suitable
onward path, and so on, until an electrical path is established between H1 and H5. The path selection
generally relies on a routing algorithm that may consider factors such as network traffic and path
length. Once a path is established, data transfer begins. There are three explicit phases involved in
circuit-switched data transfer.

Figure 3.5: Circuit switching Network

Page 6 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

Store and Forward Switching

A store and forward (S&F) network configuration is depicted in Figure 3.6.

Figure 3.6

In S&F switching, the switching nodes possess the capability to store user messages and forward
them towards the destination as soon as the links become available. For this purpose, each node is
equipped with a processor and some buffer storage. Unlike in circuit switching, no end-to-end link
is established prior to data transmission. Instead, the user deposits their message to the nearest
switching node, and from there on, the network assumes the responsibility for delivering the message
to the destination user or host. This process bears resemblance to the postal system, where letters are
collected and forwarded to their recipients.

The network moves user information from node to node, with each movement termed a hop. As
communication links are used one at a time between any two nodes, line speeds can be efficiently
utilized. S&F switching may be classified as either:

- message switching
- packet switching

In message switching, once transmission is initiated, a message is transmitted in its entirety without
interruption from one node to another. The node processor performs the following functions:

1. Receives the full user message and stores it.

2. Checks the message for data transmission errors and performs error recovery if required.
3. Determines the destination address from the user message.
4. Chooses an appropriate link towards the destination based on certain routing criteria.
5. Forwards the message to the next node on the chosen link.

Page 7 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

Message switching has certain drawbacks. For long messages, it's crucial to ensure adequate storage
space on the receiving node before transmission begins. Otherwise, buffer storage may become full,
leading to the loss of part of the message and requiring retransmission of the entire message in some
cases. Similarly, if an error occurs during transmission, the entire message may need to be
retransmitted, resulting in significant communication overheads in the network. Additionally, high-
priority short messages may be delayed if a long message is currently in transmission.

Packet switching overcomes these drawbacks by splitting messages into a number of packets, often
fixed in size, and transmitting them in a S&F fashion. Messages are split at the source host and
reassembled at the destination host. Each packet transmission is independent of the others, allowing
packets to travel via different routes and arrive at the destination with varying delays. However, this
can result in packets arriving out of sequence at the destination node.

Each packet must carry complete address information, including destination identifier (ID), source
ID, message ID, and packet ID, in addition to the actual user data. This necessitates considerable
overhead in terms of buffer storage and processing power at the network nodes. To address these
challenges, packet networks offer two different forms of services:

Figure 3.7
Page 8 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

D. DATA COMMUNICATION ARCHITECTURE

Data communication among computers involves various functions such as the physical transmission
of bits, error control, routing, and session establishment. To efficiently implement these functions,
vendors of computer systems have developed their own architectures.

Examples of vendor-specific architectures include the System Network Architecture (SNA) of IBM
and the Digital Network Architecture (DNA) of Digital Equipment Corporation (DEC). These
architectures facilitate the interconnection of computers from the same vendor but not from different
vendors. Systems or networks that are not open to other vendor systems for networking are known
as "closed" systems or networks.

To enable the interconnection of heterogeneous computer systems from different vendors as a

network, a standardized architecture that all vendors can adopt is required. Heterogeneity
encompasses various aspects such as systems from different vendors, under different managements,
of different complexities, and employing different technologies.

ARPANET, the network project supported by the Advanced Research Projects Agency of the
Department of Defense in the United States, was a pioneering effort in interconnecting heterogeneous
systems. The experience gained from this project significantly contributed to the emergence of a
world standard architecture for computer communication, largely pursued and outlined by the
International Organization for Standardization (ISO). These standards, now widely known as ISO-
Open System Interconnection (ISO-OSI) standards, are widely accepted.

The ISO-OSI standards are based on a reference architecture described in ISO standard IS 7498,
which CCITT has also adopted under its own number X.200. This architecture is considered "open"
because any vendor's system conforming to this reference model can organize information transfer
with any other vendor's system that also conforms to the same architecture.

ISO-OSI Reference Model

Before discussing this model, a few definitions are in order:

System: One or more autonomous computers and their associated software, peripherals, and users
capable of information processing and/or transfer.

Subsystem: A logically independent smaller unit of a system. A succession of subsystems makes up

a system, as shown in Figure 3.8.

Figure 3.8: Systems, subsystems and layers in ISO-OSI model

Page 9 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

Layer: A layer consists of subsystems of the same rank from all interconnected systems. This
concept is illustrated in Figure, which depicts a five-layer network. The subsystems and layers are
numbered starting with one at the bottom level.

Entity: The functions within a layer are executed by hardware subsystems and/or software packages,
known as entities. The ISO-OSI architecture follows a layered approach. Layering is a natural choice
for communication architectures, as illustrated by the following example: Consider the activities
involved when executives A and B from two companies in different cities wish to converse over a
trunk telephone connection. Assuming there is no Subscriber Trunk Dialing (STD) facility between
the two cities, let Executive A be the calling party. He requests his secretary to connect him to
Executive B. The secretary then communicates with the trunk operator, providing details such as the
calling number, called number, nature of the call, and the name of the person called. The local trunk
operator then contacts the trunk operator in the other city, relaying the details. The remote trunk
operator contacts the secretary of Executive B, who confirms with Executive B that he would like to
receive the call and requests the operator to connect the call. This process is depicted in Fig 3.9.

Figure 3.9: A three-layer structure for a trunk call connection

Key points regarding layering and communication processes:

1. A three-layer structure is utilized in this communication process.

2. Communication between upper and lower layers is strictly business-like.
3. There is generally little private conversation between trunk operators and the
secretaries due to their familiarity. Entities at the same level or layer exchange
information using their own private protocols.
4. Each layer obtains services from its immediate lower layer and provides services to
its immediate upper layer, acting both as a user and service provider.
5. Each layer has well-defined functions to perform.
6. The implementation details of each layer's functions are immaterial. For example, the
secretary may delegate the task of booking the call to an assistant, as long as the call
is booked satisfactorily.

Page 10 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

Figure 3.10: ISO-OSI reference model

Figure 3.10 illustrates the ISO-OSI reference model. Entities in these layers always communicate
with peer entities in the adjacent system. In other words, in the first three layers of the ISO-OSI
model, communication proceeds on a link-by-link basis. In contrast, entities in layers 4 - 7
communicate with peer entities in the end systems. There is no communication with entities in the
intermediate systems. In this sense, layers 4 - 7 are often called end-to-end layers.
The 7-layer architecture has been arrived at after a careful application of a broad set of layering
principles. The important principles are:

1. Create layers to handle functions that are manifestly different in the process performed or
technology involved.
2. Collect similar functions into the same layer and create a boundary at a point where the
number of interactions across the boundary is minimized.
3. Create a layer of easily localized functions so that the layer could be totally redesigned and
its protocols changed in a major way to take advantage of new advances in architectures,
hardware, and software technology without changing the services offered or the interfaces
with the adjacent layers.

E. LINK-TO-LINK LAYERS
The first three layers, namely the physical, data link, and network layers, constitute the link-to-link
layers of the OSI reference model. Entities in an OSI layer perform certain functions to fulfill the
stated purpose of the layer. They obtain services from the immediate lower layer and provide services
to the immediate upper layer. OSI services may be classified into two broad categories:

- Connection-oriented services
Page 11 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

- Connectionless services

In connection-oriented services, a connection is first established between the sender and the
receiver before data transfer can commence. The connection may be virtual (logical) or physical,
depending on the network capabilities and facilities. The essence of a connection-oriented service is
that a connection acts like a tube or a pipe, delivering the data to the receiver strictly in the same
order in which the data was put into the connection by the sender. Connection-oriented service is
modeled after the telephone system.

In contrast, connectionless service is modeled after the postal system. Each submission by the sender
is treated independently of others and is self-contained with the full address of the destination and
the source indication, which may also be the full address. In connectionless service, when two
messages are sent to the same destination one after another, it is possible that the faster one is delayed
and the second one arrives first.

Datagram service and virtual circuit service are examples of connectionless and connection-oriented
services, respectively. The operation of the two categories of services is depicted in Figures 3.11 (a)
and (b).

Figure 3.11

A connection-oriented service has provisions for acknowledgments, flow control, and error recovery,
whereas a connectionless service generally does not have such provisions. Peer entities of OSI layers
communicate using peer protocols. Protocols are strict procedures and sequences of actions to be
followed in order to achieve an orderly exchange of information among peer entities. Corresponding
to the two categories of services, there are two sets of protocols: one set for connection-oriented
services and the other for connectionless services.

It is important to recognize that layer protocols relate to the implementation of services of the layer
and therefore are not visible to users or other layers. This separation of services and protocols
provides complete freedom to change protocols at will without affecting the services.
Page 12 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

Physical Layer
It is essential that the OSI architecture permits the usage of a realistic variety of physical media and
control procedures. Keeping this in mind, the lowest layer of the architecture has been identified as
the physical layer. This layer performs functions associated with the activation and deactivation of
physical connections. It deals with encoding and decoding of signals and the bit-level transmission
of electronic signals through the available transmission medium. The transmission may be
synchronous or asynchronous. Modes of data transmission may be simplex, half duplex, or full
duplex. The physical layer provides mechanical, electrical, functional, and procedural characteristics
to activate, maintain, and deactivate physical connections for the transmission of bits.

Data Link Layer

Some physical communication media, like telephone lines, have error rates that are not acceptable
for the great majority of data network applications. Special techniques are required to ensure error-
free transmission of data.

The data link layer deals with error detection and automatic recovery procedures required when a
message is lost or corrupted. For this purpose, a user of this layer, i.e., the network layer, is required
to break up the data to be transmitted into frames, which are then numbered and transmitted
sequentially. The layer provides functional and procedural means to establish, maintain, and release
data link connections for the entities in the network layer. A data link connection may be built upon
one or several physical connections.

Another important function performed by the data link layer is the link-level flow control of frames.
Flow control is essentially a traffic regulation mechanism that must be enforced when the receiver is
unable to accept frames as fast as the transmitter is able to send. A data link may be of the point-to-
point type, as in the case of terrestrial networks, or broadcast type, as in the case of SBDNs, LANs,
or MANs. In the case of broadcast-type channels, the data link layer will have to perform an
additional function of acquiring or accessing the channel before data transmission can occur.

The main sources of error in a system are:

1. Thermal noise, which is internal to the system.

2. Impulse or spike noise, which originates from man-made sources like automobiles or
signaling in telephone systems, and from natural sources like lightning.
3. Crosstalk, which occurs through electromagnetic radiation from parallel and adjacent wires
behaving like an antenna.

Errors are of two types: isolated and bursty. Errors due to thermal noise are generally isolated,
whereas errors due to spike noise or crosstalk are bursty in nature. Error control mechanisms are
chosen depending on the type of error that is predominant in a given system. Three error control
mechanisms are commonly used:

1. Echo checking.
2. Forward error correction (FEC).
3. Automatic repeat request (ARQ).

Network Layer
The highest link-to-link layer in the OSI model is the network layer. Although this layer functions
on a link-to-link basis, it is concerned with the transmission of packets from the source node to the
Page 13 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

destination node. It deals with routing and switching considerations necessary for establishing a
network connection, which may involve the use of several transmission resources in tandem,
including a number of intermediate switching nodes of different subnetworks.

Figure 3.12: Classification of routing algorithms

The network layer abstracts the details of the underlying communication media and the different
characteristics of the transmission and network technologies from the transport layer. Its role is to
ensure a certain quality of service to the upper layers. Since an end-to-end connection may involve
routing through various networks, internetworking becomes a vital function of the network layer.
Internetworking involves addressing schemes, network capabilities, protocol differences, and
accounting and billing. Additionally, the network layer tackles network congestion, which may occur
due to an excessive number of messages on a particular route.

Several measures can be used to assess the performance of a routing algorithm:

1. Minimum delay
2. Minimum number of intermediate nodes or hops
3. Processing complexity
4. Signaling capacity required on the network
5. The rate of adaption in the case of adaptive algorithms
6. Fairness to all types of traffic
7. A reasonable response time over a range of traffic intensities
8. Robustness: the ability to reach the destination even when parts of the network fail
9. Stability: the ability to reach the destination quickly without wandering.

F. END-TO-END LAYERS
Many data networks were operational before the OSI model was designed. These networks were well
thought out up to the network layer, but little had been done about the transport layer and above.
Consequently, the design of the bottom three layers of OSI was highly influenced by pre-OSI
developments.

Transport Layer
The transport layer is the first end-to-end layer in the OSI architecture. It is responsible for matching
user message characteristics and service requirements with those of the network capabilities. In a
Page 14 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

packet-switched network, the transport entity breaks up a long user message into packets to match
the network capabilities. The packets are reassembled at the destination transport subsystem to
reconstruct the user message. The transport entities may invoke sophisticated error control protocols
to provide a reliable session service on an unreliable network. Similarly, a number of low-rate user
services may be multiplexed to efficiently use a single network connection, or a high data rate
requirement from the user may be split into several network connections. Multiplexing and splitting,
performed by the transport layer, are transparent to the session layer. Like every other layer, the
transport layer is also concerned with the establishment, control, and release of the transport
connections between peer entities in the source and destination systems. End-to-end flow control and
error recovery are also functions of the transport layer. The need for flow control arises when the
speed or the buffer space of the destination machine does not match that of the source machine.
Similarly, end-to-end error recovery becomes necessary when the source or destination system fails
or when the network becomes disconnected due to link failures.

As far as a user is concerned, it is the transport layer that offers transport services regardless of the
underlying network. The user makes service requests to this layer by specifying certain "quality of
service" (QoS) parameter values. Some QoS parameters that are of direct interest to the users are:

1. Transit delay
2. Residual error rate
3. Protection
4. Transfer failure probability
5. Priority
6. Throughput.

Session Layer
The main function of the session layer is to organize different sessions between cooperating entities
and perform all related functions like synchronization, failure management, control, etc., for the
successful execution of a session. Online searches of databases, remote job entry, remote login to a
time-sharing system, and file transfer between two systems are all examples of sessions. Different
sessions have different requirements. For example, a dialogue session may be two-way simultaneous
or one-way alternate. A large file transfer session may call for rollback points to be established to
recover from system crashes. An online transaction processing session calls for semaphore
management, file, record, and sometimes even item-level lock mechanisms. A quarantine service
enables a specified number of presentation layer SDUs to be transported to the destination system
but not delivered unless explicitly requested by the sender. The session layer may also offer a
directory service.

Figure 3.13: Synchronization points in session layer

Page 15 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

Presentation Layer
The purpose of the presentation layer is to represent information to the communicating application
entities in a way that preserves meaning while resolving syntax differences. Syntax differences are
resolved by encoding application data into a standard abstract notation valid throughout the network.
Thus, file format differences (e.g., IBM or DEC format), data representation differences (e.g., ASCII
or EBCDIC), or data structure differences are all resolved using a standard notation. Data
transformation and formatting may include data compression, encryption, etc. There are two aspects
associated with network-wide handling of a variety of data in a standard form: the representation of
data in a standard form and the transmission of the data as a bit stream across the network.

Identifier Length Data Flag

Figure 3.14: ASN.1 transfer syntax

1. Universal
2. Application
3. Private
4. Context specific

Application Layer
As the highest layer in the OSI reference model, the application layer provides services to the users
of the OSI environment. The layer provides all services that are directly comprehensible by the users,
including:

1. Electronic mail or message handling services

2. Directory services
3. Cost allocation
4. Determination of quality of service
5. File transfer and management
6. Editors and terminal support services
7. Telematic services like videotex.

Figure 3.15: Virtual terminal data structure models

Page 16 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

In general, every application requires its own software, which, in turn, uses a number of generic
supporting packages. These generic packages and the application-specific packages are part and
parcel of the application layer. The list above consists of generic applications and support packages.
For example, file transfer or remote file access may be used by airline reservation systems, banking
applications, etc. Similarly, electronic mail may be used in various contexts.

G. SATELLITE-BASED DATA NETWORKS

In the context of data networks and the OSI reference model, several important aspects of satellite
communication systems require consideration:

1. Satellite network topology, configurations, modulation schemes, and bandwidth utilization:

These are aspects related to the physical layer functions of the reference model.
2. Being a common communication resource accessible by all or a group of earth stations
simultaneously, media access becomes a nontrivial function in the data link layer.
3. Satellite communication, being broadcast in nature, makes routing a trivial function;
however, organizing point-to-point or point-to-multipoint connections in a broadcast.
4. Since a geostationary communication satellite is placed at an altitude of about 36,000 km
above the equator, the signal will have to travel a distance of 72,000 km or more between the
source and the destination, resulting in a significant propagation delay of 250–300 ms. The
session and transport layers have to be concerned about this factor.

LAN
A local area network (LAN) exemplifies a distributed environment and finds applications in various
areas. Some examples include:

1. Office automation
2. Factory automation
3. Distributed computing
4. Fire and security systems
5. Process control
6. Document distribution.

Of these, the first two applications have evolved to the extent that world standards are emerging for
them. The office automation standard, known as Technical and Office Protocol (TOP), and the
factory automation standard, known as Manufacturing Automation Protocol (MAP), are discussed
in Section 10.11. Being distributed in nature, LANs offer numerous benefits, but they also suffer
from certain disadvantages.

Advantages of LANs include:

1. Flexibility: Unlike a large centralized system, a LAN can evolve over time. It can be put into
operation with a small investment, and more systems can be added as needed.
2. Redundancy: Since a LAN consists of multiple interconnected systems, it offers good backup
capability in case one or two systems fail, enhancing reliability and availability to users.
3. Resource Sharing: LANs provide a resource-sharing environment where expensive
peripherals, hosts, and databases can be shared by all users.
4. Vendor Neutrality: A LAN adhering to a certain standard allows multivendor systems to be
connected, reducing dependency on a single vendor.

Page 17 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

5. Compatibility: LANs typically support various types of systems, allowing for interoperability
and compatibility among different devices.

Three access methods are predominantly used in LANs: switched access, contention or multiple
access, and token passing access. Switched access is common in LANs designed around CBXs, while
contention and token passing access techniques are also widely used.

Different types of LANs are obtained by choosing combinations of medium, topology, and access
method. Common combinations include multiple access bus, token passing ring or token ring, and
token passing bus or token bus.

METROPOLITAN AREA NETWORKS (MANs)

A metropolitan area network (MAN) usually covers a geographical area spanning 5–50 km. MANs
support services that require guaranteed bandwidth and bounded delay performance, in addition to
data services without such restrictions. Additional services may include voice and video, depending
on available bandwidth. MANs may operate at speeds of 1 Mbps and above, with network structures
similar to those used in LANs: star, bus, and ring.

FIBER OPTIC NETWORKS (FONs)

Fiber optic networks are characterized by:

1. High-speed operation (typically 100 Mbps or more)

2. Ability to span large distances (100–200 km)
3. Support for a moderate number of stations (typically 10 to a few hundred, with a
maximum limit around 1000).

Fiber networks may be configured around star, ring, or bus structures.

Figure 3.16
Page 18 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

The number of stations that can be supported in a star or a bus structure is relatively low compared
to that in a ring configuration. Optical fibres are inherently unidirectional and this influences the way
in which the network structures are realized. Medium access considerations for fiber optic networks
are similar to those for MANs.

The central hub in a star FON may be passive or active. The passive hub typically consists of a silica
cylinder, with incoming and outgoing fibers fused to each end. Limited stations in passive-star FONs
permit a simple medium access protocol to be used, where each station is assigned one or two time
slots for transmission.

DATA NETWORK STANDARDS

Three major international bodies significantly contribute to data network standards:

4. International Organization for Standardization (ISO)

5. International Telecommunication Union - Telecommunication Standardization Sector
(ITU-T)
6. Institute of Electrical and Electronics Engineers (IEEE).

ISO and ITU-T standards have international legal standing, while IEEE standards may need adoption
by ISO for similar legal standing. ANSI sponsors IEEE standards work, which is affiliated with ISO.
Standards evolved by ITU-T are adopted by ISO and vice versa, with minor changes introduced
during adoption. For example, ISO's OSI reference model (ISO 7498) is adopted by ITU-T as X.200.
Similarly, ITU-T's X.400 message handling system is adopted by ISO as the Message Oriented Text
Interchange System (MOTIS) under IS 10021.

Page 19 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

QUESTIONS
1. EPABX Systems:

- What is the basic block diagram of an EPABX system?

- What are the key features and functionalities of EPABX systems?
- How do EPABX systems work in managing internal and external calls within an
organization?

2. Data Transmission in PSTNs:

- How is data transmitted over Public Switched Telephone Networks (PSTNs)?

- What are the protocols and technologies involved in data transmission over PSTNs?
- What factors affect the efficiency and speed of data transmission in PSTNs?

3. Data Rates in PSTNs:

- What are the typical data rates supported by PSTNs for data transmission?
- How do data rates vary for different types of connections within PSTNs?
- What factors influence the choice of data rates in PSTNs?

4. Modems:

- What is a modem, and what is its role in data communication?

- How do modems convert digital data to analog signals for transmission over PSTNs?
- What are the different types of modems and their respective functionalities?

5. Switching Techniques for Data Transmission:

- What are the different switching techniques used for data transmission?
- How does circuit switching work in the context of data communication?
- What is store-and-forward switching, and how does it differ from other switching techniques?

6. ISO-OSI Reference Model:

- What is the ISO-OSI reference model, and what is its purpose?

- Can you describe the different layers of the OSI model and their functions?
- How does the OSI model facilitate interoperability between different networking
technologies?

7. Satellite-Based Data Networks:

- What are satellite-based data networks, and how do they operate?

- What are the advantages and disadvantages of using satellites for data transmission?
- Can you provide examples of applications that benefit from satellite-based data networks?

8. LAN (Local Area Network):

- What is a LAN, and what are its primary characteristics?

- How do LANs facilitate local data communication within an organization or geographical
area?
- What are some common topologies and protocols used in LANs?

Page 20 of 21
BEE4102 TELEPHONE SYSTEMS CHAPTER 3

9. Metropolitan Area Network (MAN):

- What distinguishes a MAN from other types of networks, such as LANs and WANs?
- How are MANs structured, and what are their typical applications?
- Can you provide examples of technologies used in MANs?

10. Fiber Optic Networks:

- What are fiber optic networks, and how do they differ from traditional copper-based
networks?
- What are the advantages of using fiber optics for data transmission?
- How are fiber optic networks deployed and maintained?

11. Data Network Standards:

- What role do standards play in data networking?

- Can you provide examples of key data network standards and their importance?
- How do standards bodies ensure interoperability and compatibility among different
networking technologies?

Page 21 of 21