UNIT 3 Computer Network and Communication

 

UNIT 3

 

Computer Network
and Communication

Communication
Network

Ever since the
beginning of time, people have had an uncontrollable need to communicate. Our
nature drives us to exchange ideas, information and opinions. Without
communication, we can’t learn, without learning, we can’t grow. and without
growth, we shrivel up and disappear.

Creativity and
communication drive the human race more than any other force. They are the
essential elements of life.

Networking is the
ultimate level of communication. It transcends words and pictures to provide
the pathway for thoughts, ideas and dreams. Networking that exists today is the
result of millions of years of evolution and growth.

If we go through the
history of communication, Charles Babbage and Countess of Lovelace are listed
as the pioneer in the field of the present communicating world. Suddenly, the
alphabet shrunk from twenty-six characters to two i.e. 0 and 1.

The world was thrust
into a technological era that would spawn numerous networking inventions
including the telephone, phonograph, motion pictures and the television.
Finally, data communications were born a century later when computers were
linked across distances.

NASA uses data
communications to control the space shuttle and realign geosynchronous
satellites, surgeons perform computerized operations from miles away, and
because of data communications, today’s workforce is better to balance home and
work by working from home or telecommuting. Data communications have had a
profound effect on humanity.

The world will go on
changing. It will become a small global village. In the new information age,
humans will become very big fish in a very small pond.

Computer
Network

A more comprehensive
definition of the network: “A collection of distributed, intelligent
machines that share data and information through interconnected lines of
communication is called network”.

When two or more
computers are brought together in connection with cable or without cable, that
may extend within a limited room or to the entire world, then the computers are
said to be in network connection. They merely perform the task of data sharing,
transmission of data and communication.

Network
Elements

Network elements fall
into a place to create a compact LAN. Network communications provide the
electronic pathway that computers use to communicate with each other. These
elements include physical cabling, data signals and LAN standards.

Networking represents
the rules of the road, defining protocols that instruct different machines how
to speak (Internetwork Packet Exchange/Sequence Packet Exchange) for NetWare,
TCP/IP (Transmission Control Protocol/Internet Protocol) for the Internet and
AppleTalk for Macintosh computers.

Finally, network
services simply define the things networks can do. Service requesters (clients)
ask for resources and service providers (servers) grant them in the case of
Client Server architecture. Common network services include file services,
print services and messaging. A service provider is any combination of hardware
and software that fulfills a particular job or function.

Network elements
define the pathway and protocols for electronic communications. We are more
concerned with what the message look like and how it travels from point A to
point B in the imaginary computers in the network.

Here are three basic
elements of networks:

1.    
Network service

2.    
Network Standards & Protocols

3.    
Transmission media

Network
Services

Network services are
the things that a network can do. These are the services provided by the
computer. The major services that a network can provide are:

1.    
File Services

2.    
Print Services

3.    
Message Service

4.    
Application Services

5.    
Database Services

1. File Service: File Services
includes file transfer and storage, data migration, file update synchronization
and archiving. These features are the most popular reasons for networking.

2.Print-Service:

Print Services produce shared access to valuable printing devices. A printer
that is used as sharing in the network acts as the print server.

3.Message-Service:

Message Services facilitate e-mail, manage integrated email and voicemail and co-ordinate
object-oriented applications. Messaging is an exciting new player at the
network party.

4.Application-Service:

Application Services allows you to centralize high-profile applications. The
server holds the applications (generally network versions) and they are
accessed by the users in the network.

5.Database-Service:

Database Services involve the coordination of distributed data and replication.
The main server holds the data and determines the number of users that can
access the database and modify the data. It also provides the print facility of
the data in the network.

Network
Standards & Protocols

The term protocol
refers to the set of rules and procedures that govern the transmission of
messages over a physical networking medium. Topology describes the geographic
orientation and arrangement of networking components.

Together, protocol
and topology combine to create a networking standard. These standards are
developed and controlled by the Institute of Electrical and Electronics
Engineers (IEEE). The four major industry standards are:

1.    
Ethernet

2.    
Token Ring

3.    
ARCnet and

4.    
FDDI (Fiber Distributed Data Interface)

Topologies define the
geographic arrangement of server and workstations. File servers and
workstations are arranged according to a variety of factors: speed, cost,
reliability, distance and load requirements. Each topology is ideally used for
different combinations of these factors:

·        
The bus is fast, cheap and unreliable.

·        
The star is slow, large and reliable.

·        
The ring is fast, expensive and reliable.

·        
The starring is fast, cheaper and reliable.

Ethernet Standard for
Bus Topology

The Ethernet topology
was developed at the University of Hawaii, to connect computers on the various
islands. It was a radio-based design. The term Ethernet is a compound word of
either meaning air and net meaning network. Later, Robert Metcalfe went to
Xerox’s Palo Alto Research Center (PARC) laboratories.

The radio portion was
eliminated and changed to co-axial cabling. This was the string from station to
station. This station to station topology was named linear bus.

Token-Passing
Standard or Protocol

The token-passing
protocol relies on a control signal called the token. A token is a 24-bit
packet that circulates throughout the network, from NIC to NIC, in an orderly
fashion. If a workstation wants to transmit a message, first, it must seize the
token. At that point, the workstation has complete control over the
communications channel.

The existence of only
one token eliminates the possibility of signal collisions. This means that the
only station can speak at a time on a multi-station ring.

ARC Net Standard or
Protocol

The Attached Resource Computer Network (ARC
Net Standard was created in 1977, at the Data point Corporation by a scientist
named John Murphy. This standard predates the token-passing design adopted by
IBM but uses very much the same technology ARC Net does not offer the same
overall connectivity as Ethernet, but its hardware components are standardized
sufficiently so that any ARC Net device from any manufacturer can be used on
any other ARC Net LAN.

 

Direction
of communication flow

Local Area Networks
(LANs)have some specific architecture for communicating among the computers in
the connection. The networks made by connecting IBM PCs and Macintosh computers
have different methods of communication between themselves. The architecture
defines how the computers communicate in these networks. The different
architectures are:

1.    
Peer-to-Peer Architecture

2.    
Client Server Architecture

Peer-to-Peer
Architecture

Some of today’s most
popular server-centric operating systems include Novell’s NetWare, Microsoft’s
Windows NT Advanced Server, Windows 2000 Advanced Server, IBM’s LAN Server and
Banyan’s VINES. Not every network is server-centric.

Some system called
peer-to-peer network software usually has no central point of control, relying
on the peers to perform the various functions for each other. Some of today’s
most popular peer-to-peer operating systems includes Microsoft Windows 95,
Microsoft Windows for Workgroups, Microsoft Windows Nt, LAN tactics and
Novell’s Personal NetWare.

 

In a peer-to-peer
architecture, there is no dedicated server. The data signal sent by a computer
is recognized by another computer and it is accessed by using intelligent
methods. In this case, every computer can act as a server as well as the
client. The computer that provides servers is a server and the computer that
receives the information is a client at the time of data transmission and
communication.

Client
Server Architecture

In the LAN, the
computers are confined with the limited boundary of 1 to5 km by using
repeaters. The data transmission is the rate of 20 MBPS (million bits per
second). In this case, a dedicated server is installed in a computer and all
the others computers are clients.

 

The operating system,
applications and other resources are generally kept in the file server only. A
server can handle 20-50 computers ( depends on the server capacity). All the
computers are connected by using co-axial cable. The operating systems used in
client / server is the computer that provides the services and the computers
that requests data, program and the query is known as a client computer.

Types of
Networks: Computer Network Models

Network models
classify networks by size, distance and structure. These models help us to
understand the relationship among centralized, distributed and collaborative
computing systems. There are three fundamental classifications of network
models:

1.    
Local Area Networks (LANs)

2.    
Metropolitan Area Networks (MANs)

3.    
Wide Area Network (WANs)

Local Area
Network (LAN)

A local area network
combines computers hardware and transmission media in the relatively small
area. These systems are usually contained within one department, building or
campus. In addition, LANs typically comprise only one transmission media type
such as coaxial cable or a twisted pair but never both.

LANs are
characterized by comparatively high-speed communications. These high speeds are
possible because LANs use one kind of cable, which generally is limited to 5 km
or less. MANs and WANs, on the other hand, typically rely on inbound
communication media which equates to much slower speeds.

LANs occupy a
relatively small area, including a relatively small number of machines and all
components connect directly to the communications media. LAN hardware gives the
system its processing communications, system fault tolerance and most
important, connectivity. LAN software endows the system with productivity, user
interface, network management and most important, user transparency. The two
most important components are LAN are the file server (hardware) and network
operating system ( software).

Characteristics of
LAN

The following are the
major characteristics of LAN which are physically connected.

1.    
The distance between two nodes is not more than 5 km.

2.    
LAN is confined within the same management system.

3.    
The data transfer rate ranges from 1 byte to 10 GB per second.

4.    
This is used in office automation system

5.    
There exists node-to-node connection.

Capacities of LAN

The following are the
major capacities of LAN in addition to other capacities which are not discussed
here.

1.    
LAN helps to access distance data, processes them and prints.

2.    
Shares hardware resources and software.

3.    
It can control and transmit data on the basis of priority.

4.    
Acts and bridge between two networks.

5.    
Acts as a gateway for internetwork communication.

Metropolitan
Area Network (MAN)

A Metropolitan Area
Network (MAN) is larger than a LAN, as its name implies, it covers the area of
a single city. MANs rarely extend beyond 100 KM and frequently comprise a
combination of different hardware and transmission media. In addition, they
have somewhat slower data communication rates than that of LAN’s, partly
because of their reliance on unbound media over great distances.

The most important
characteristics of MAN’s are their diversity. MAN’s typically are used when you
need to connect dissimilar systems within a single metropolis. As shown in
figure, we have connected bank, home and work LAN’s into one MAN. This level of
continuity is synonymous with twenty-first-century computing.

The two most
important components of MAN’s are security and standardization. Security is
important because information is being shared between similar systems. Many
users are still wary of MAN computing because of the sensitivity of their data.
Standardization is necessary to ensure reliable data communications.

Wide Area
Network (WAN)

A Wide Area Network
(WAN) is simply a LAN of LANs. WAN connect LANs that may be on opposite sides
of a building, across the country or around the world. On one of the three
network models, WAN’s are characterized by the slowest data communication rates
and the largest distances. Furthermore, WANs can connect different types of
networks together.

For example, you can
connect a centralized computing system (mainframe) with a distributed LAN.
These connections frequently are made through a special type of device called a
gateway. WAN connects LANs from around the world. WANs can be characterized as
either enterprise or global WANs.

WAN connects LANs
from around the world. WANs can be characterized as either enterprise or global
WANs. An enterprise WAN connects an entire organization, including all LANs at
various sites. This term is used for large, widespread organization such as
corporations, universities and governments. Even though the LANs may be in
different part of the country; they must all belong to a single company or
institution.

Global WANs also span
the world, but they do not have to connect LANs within a single organization.
The internet is an example of a global WAN. It connects diverse locations,
organizations and institutions throughout the world. Finally, global WANs can
be public or private. The Internet, for example, is a public global network,
which means that anyone can attach to it from anywhere in the world. Private
WANs (intranet) are also available.

As more networks are connected together, the
current classifications may begin to disappear. Technological progress is
expected to result in the development of a single computer networking
infrastructure. Faster and cheaper data communications will transcend the need
for LANs, MANs and WANs.

 

LAN
Topologies

LAN physical topology
defines the geographical arrangement of networking devices. Topologies are
driven fundamentally by two network connection types:

A point-to-point
connection is a direct link between two devices. For example, when you attach
your computer to a printer, you have created a point-to-point link. In
networking terms, most of the today’s point-to-point connections are associated
with modems and PSTN (Public Switched Telephone Network) communications because
only two devices share point-to-point connections, it defeats the purpose of a
shared network.

A multipoint
connection, on the other hand, is a link between three or more devices.
Historically, multipoint connections were used to attach central CPUs to
distributed dumb terminals.

In today’s LAN
environments, multipoint connections link many network devices in various
configurations.

The major topologies
of LAN are:

1.    
Bus Topology

2.    
Ring Topology

3.    
Star Topology

4.    
Mesh Topology

5.    
Cellular Topology

6.    
Hybrid Topology

Bus
Topology

The physical bus
topology is the simplest and most widely used of the network designs. It
consists of one continuous length of cabling (trunk) and a terminating resistor
(terminator) at each end. The data communications message travels along the bus
in both directions until it is picked up by a workstation or server NIC.

If the message is
missed or not recognized, it reaches the end of the cabling and dissipates at
the terminator. All nodes in the bus topology have equal access to the trunk –
no discriminating here. This is accomplished using short drop cables or direct
T-connectors.

This design is easy
to install because the backbone trunk traverses the LAN as one cable segment.
This minimizes the amount of transmission media required. Also, the number of
devices and length of the trunk can be easily expanded.

Advantages of Bus
Topology:

1.    
It uses established standards and it is relatively easy to
install.

2.    
Requires fewer media than other topologies.

Disadvantages of Bus
Topology:

1.    
The bus networks are difficult to reconfigure, especially when
the acceptable number of connections or maximum distances have been reached.

2.    
They are also difficult to troubleshoot because everything
happens on a single media segment. This can have dangerous consequences because
any break in the cabling brings the network to its knees.

 

 

Ring Topology

As its name implies,
the physical ring topology is a circular loop of point-to-point links. Each
device connects directly or indirectly to the ring through an interface device
or drop cable. Messages travel around the ring from node to node in very
organized manner. Each workstation checks the messages for a matching
destination address.

If the address
doesn’t match, the node simply regenerates the message and sends it on its way.
If the address matches, the node accepts the message and sends a reply to the
originating sender. Initially, ring topologies are moderately simple to
install; however, they require more media than bus systems because the loop
must be closed.

Once your ring has
been installed, it’s a bit more difficult to reconfigure. Ring segments must be
divided or replaced every time they’re changed. Moreover, any break in the loop
can affect all devices on the network.

Advantages of Ring
Topology:

1.    
They are very easy to troubleshoot because each device
incorporates a repeater.

2.    
A special internal feature called becoming, allows the troubled
workstation to identify themselves quickly.

Disadvantages of Ring
Topology:

1.    
It is considerably difficult to install and reconfigure ring
topology.

2.    
Media failure on unidirectional or single loop causes complete
network failure.

Star
Topology

The Physical star
topology uses a central controlling hub with dedicated legs pointing in all
directions – like points of a star. Each network devices has a dedicated
point-to-point link to the central hub. This strategy prevents troublesome
collisions and keeps the line of communication open and free of traffic.

Star topologies are
somewhat difficult to install because each device gets its own dedicated
segment. Obviously, they require a great deal of cabling. This design provides
an excellent platform for reconfiguration and troubleshooting.

Changes to the
network are as simple as plugging another segment into the hub. In addition, a
break in the LAN is easy to isolate and doesn’t affect the rest of the network.

Advantages of Star
Topology:

1.    
Relatively easy to configure.

2.    
Easy to troubleshoot.

3.    
Media faults are automatically isolated to the failed segment.

Disadvantages of Star
Topology:

1.    
Requires more cable than most topologies.

2.    
Moderately difficult to install.’

Mesh Topology

The mesh topology is
the only true point-to-point design. It uses a dedicated link between every
device on the network. This design is not very practical because of its
excessive waste of transmission media. This topology is difficult to install and
reconfigure.

Moreover, as the
number of devices increases geometrically, the speed of communication also
become slow. ATM (Asynchronous Transfer Mode) and switched Hubs are the example
of high-speed Mesh implementation.

Advantages of Mesh
Topology:

1.    
Easy to troubleshoot because each link is independent of all
others.

2.    
You can easily identify faults and isolate the affected links.
Because of the high number of redundant paths, multiple links can fail before
the failure affects any network device.

Disadvantages of Mesh
Topology:

1.    
It is difficult to install and reconfigure especially as the
number of devices increases.

Cellular
Topology

A cellular topology
combines wireless point-to-point and multipoint designs to divide a geographic
area into cells. Each cell represents the portion of the total network area in
which a specific connection operates. Devices within the cell communicate with
a central station or hub. Hubs are then interconnected to route data between
cells.The cellular topology relies on the location of wireless media hubs.
Cellular networks exhibit interesting characteristics since this topology do
not depend on cables. Troubleshooting is easy because each hub interacts
independently with each device. A cellular installation depends on the accessibility
hub locations.

Advantages of
Cellular Topology:

1.    
It is relatively easy to install.

2.    
It does not require media reconfiguration when adding or
removing users.

3.    
Fault isolation and troubleshooting is fairly simple.

Disadvantages of
Cellular Topology:

1.    
All devices using a particular hub are affected by a hub
failure.

Hybrid
Topology

By modifying or
combining some of the characteristics of the ‘pure’ network topologies, a more
useful result may be obtained. These combinations are called hybrid topologies.
Some of the hybrid topologies are:

1.    
Tree network

2.                
Star-Ring or interconnected

 

Communication
Transmission Media

INTRODUCTION
T0 TRANSMISSION MEDIA

Transmission media
provide the physical path through which electrons flow. At the physical layer
of OSI model of networking, electrons represent network data as binary 0s and
1s. Transmission media provides these electrons with a bound or unbound communications
path. Telephone lines are well proven and commonly used communication media.

Basically, there are
two methods by which the communication takes place. They are:

1.    
Wired Transmission Media or Bound Transmission Media and

2.    
Wireless Transmission Media or Unbound Transmission Media

Bound
Transmission Media

As said earlier,
bound transmission media consists of a central conductor surrounded by a
physical jacket. This property offers advantages in security, reliability and
speed. Bound media are ideal for LANs but distance limitations can be the
problem for WANs. The four types of bound media are:

i. Unshielded Twisted
Pair (UTP)

ii. Shielded Twisted Pair (STP)

iii. Coaxial Cable

iv. Optical Fiber

i. Unshielded Twisted
Pair (UTP)

UTP is the most
common type of telecommunication media these days. It is most suited for data
and voice communication. It consists of two metal conductors (usually copper)
hat which are insulated separately with their own colored plastic insulation.
It can transmit up to 96000 bps. UTP usually is intended for analog
communications not digital.

ii. Shielded Twisted
Pair (STP)

STP cable has a metal
foil or braided mesh covering that covers each pair of insulated conductors.
The metal foil is used to prevent infiltration of electromagnetic noise. This
shield also helps to eliminate crosstalk during the telephone conversation. STP
provides shielded protection against EMI (Electromagnetic interference) whereas
UTP doesn’t have.

iii. Coaxial Cable

Coaxial cabling is
commonly used for television transmissions. It provides higher bandwidth and
better reliability. Unlike twisted pairs that have two wires, coaxial cables
have the single central conductor, which is made up of solid wire. This
conductor is surrounded by an outer jacket made up of PVC. A coaxial cable is
capable of transmitting data at the rate of 10 Mbps.

iv. Optical Fiber

Fiber optics plays by
a completely different set of rules. You won’t find any electricity here.
Instead, fiber-optic cabling uses pulses of light (photos) for network
communications i.e. it relies on photonics instead of electronics. As a result,
fiber optics is completely immune to EMI (Electromagnetic Interference) and is
extremely fast.

Analog and Digital
Signal

The major role of the
physical medium is to move information from one communicating device to
another. However, information to be transmitted should be first transformed
into electromagnetic signals. Information over any medium is transmitted by two
methods.

 

They are:

i. Analog Signal

ii. Digital Signal

i. Analog Signal

An analog signal is a
continuous waveform that changes smoothly over time. The analog signals
represent the continuous variation in the data. When data are plotted on the
Y-axis with time, we get curves called analog signal. During data transmission,
the signals are modulated or changed in analog form.

ii. Digital Signal

Digital Signal data
is the data stored in the form of 0s and 1s. When the signal is at a high
point, its value is 1 and when it is low, its value is 0. A signal in digital
format has precise voltages that are not affected by noise or attenuation
compared to analog signals, which are very prone to noise. Digital data
represent two states either 0 or 1.

Unbound
Transmission Media or Wireless Communication

Unbound transmission
media extend beyond the limiting confines of cabling. They provide an excellent
communication alternative for WANs. The lack of physical restrictions provides
larger bandwidth as well as wide-area capabilities. Unbound media typically operate
at very high frequencies. The four types of unbound media are:

1.    
Radio Wave

2.    
Microwave

3.    
Infrared

4.    
Satellite Communication

1. Radio Wave

Although radio waves
are prevalent and well understood, we are just beginning to realize their
enormous potential as a networking medium. Radio waves can operate on single or
multiple frequency bands. In this case, the signals are carried over carrier
waves which have frequencies in the range of radio frequency spectrum.

There are three types
of RF (radio frequency) propagation, namely, ground wave, ionosphere and line
of sight. During the time of radio wave propagation, it should be modulated.
The following are the different types of modulation.

1.    
Amplitude Modulation (AM)

2.    
Frequency Modulation (FM)

3.    
Phase Modulation (PM)

2. Microwave

Microwave, in
contrast, have been used in data communications for a long time. They have a
higher frequency than radio wave and therefore, they can handle larger amounts
of data. There are, of course, problems with microwaves attenuation and environmental
interference.

Microwave
transmission is the line of sight transmission. The transmit station must be in
visible contact with the receive station. This sets the limit on the distance
between stations depending on the local geography.

3. Infrared

Infrared offers a
great unbound photonic solution. Like fiber-optic cabling, infrared
communications use light. So, they are not bound by the limitations of
electricity.

4. Satellite
Communication

Satellite
communication is also kind of line of sight transmission. Satellites are set in
geostationary orbits directly over the equator, which rotates in
synchronization to the earth and hence looks stationary from any point on the
earth. These geostationary orbits are placed 36,000km above the earth’s
surface.The communication is carried through uplinks and downlinks. Uplinks and
downlinks are also called earth stations because they are located on the earth.
The area shadowed by the satellite in which the information or data can be
transmitted and received is called footprints. There are many satellites in the
space for communication.

Computing
Models

It is better to
appreciate the importance of network models from the evolution of computing. In
the beginning, electronic computers were huge machines that used thousands of
vacuum tubes. This was the dark ages of computing before transistors and
silicon chips. Computing, in its earliest form, was centralized in the
mainframes. Mainframe computers stored, organized and processed information.

Data were fed into
these centralized machines through remote terminals and keyboards. All data
were stored on the mainframe and processing was performed there as well.

In a distributed
computing model, 95% of the processing is handled by distributing computers.
The central file server handles housekeeping tasks only, such as security and
printer sharing. A relatively new trend called collaborative computing has
emerged largely because of distributed computing and the capabilities of
personal computer networks.

Unlike the
distributed computing model (in which individual PCs carry out independent
tasks), collaborative computing involves two or more computers working together
synergistically.

Today’s networks integrate personal
computers, mainframes and a host of other computing and other communication
services. To classify this complexity, you need network models that transcend
these definitions.

 

Network
Transmission Devices

INTRODUCTION

Network transmission
devices connect individual workstations and servers into a synergistic LAN.
These devices start simple connectors and evolve in complexity and
sophistication. The following are the transmission devices:

1.    
Media Connectors

2.    
Network Interface Card (NIC)

3.    
Repeaters

4.    
Hubs

Media
Connectors

Transmission media
connectors attach directly to the medium itself and serve as the physical
interface between cabling and network nodes. BNC T-connector, RJ- 45 Connector
for UTP, IBM data connector for STP and DIX connector for thick coaxial cabling
are media connectors. Some of the media connector are described below:

·        
BNC T-connector for Thin Coaxial Cabling

The BNC connector was
one of the first for network communications. It was introduced so long ago that
none is really sure what the letters stand for. Some authorities claim that it
means “British Naval Connector”, whereas others insist that it stands for
“Bayonet Nut Coupler”. BNC has become part of standard networking language.

·        
RJ-45 Connector for UTP (Unshielded Twisted Pair)

In the business
world, coaxial cabling is one way out, but manufacturers are still using it.
Nearly, all new cable installations use twisted-pair cabling with eight wires
per cable. The eight wires are separated into four twisted pairs.

The pairs are then
terminated in and RJ-45 connector. This port looks like the one attached to
your home telephone except that it is a bit larger with room for eight for
wires instead of four.

·        
DIX Connector for Thick Coaxial Cabling

DIX is an acronym for
Digital, Intel and Xerox. The companies that invented Ethernet these early
connectors are D-shell shaped with 15 pins each (organized in two rows and
eight pins, respectively). The drop cable that plugs into this port looks a lot
like your video connector except that it is little longer and narrower.

Network
Interface Boards or Network Interface Card (NIC)

The NIC contains the
electronic circuitry needed to ensure the reliable communications between
workstations and servers. The NIC is the electronic interface between the
computer and the LAN cabling.

The card itself uses
a bus-specific edge connector to plug into the computer motherboard. On the
exposed side of the NIC, there are cabling ports that plug directly into the
LAN cabling.

Repeaters

As the name implies,
repeaters repeat network data. In general, repeaters operate at the electronic
level and contain no real intelligence. A repeater accepts weak signals,
electrically regenerates them and then sends the messages on their way. There
are two types of repeaters: amplifiers and signal-regenerating repeaters.

Hubs

Technically speaking,
a hub is simply a multi-port repeater. In addition to regenerating network
data, hubs add form and function to the layout of the LAN. In many topologies,
the hub is the central component of the network transmission media. There are
three types of hubs. They are the passive hub, active hub and intelligent hub.

Internetwork
Transmission Devices

A LAN links a small
group of functionally similar workstations within a local geographic area. Once
your network expands to include other floors or even other LANs, it ceases to
be local and becomes a WAN. To connect multiple LANs into a WAN, you need
flexible advanced internetwork transmission devices such as:

1.    
Bridges

2.    
Routers

3.    
Gateways

1. Bridges

Like repeaters,
bridges extend the maximum distance of your network by connecting separate
segments together. However, unlike repeaters, bridges do it intelligently.
Bridges analyze network packets from multiple segments and determine who gets
through and who doesn’t. This is based on physical address.

2. Routers

Routers are more
intelligent than bridges. Instead of being limited to the physical address,
routers can use both physical and logical addressing, which allows you to break
your internet work into logical subnets.

3. Gateways

Gateways represent
the pinnacle of internetwork transmission devices. They perform more software
translation than anything. Routing capabilities which are built into the
gateway device do the packet routing. Gateways are required when network
messages travel between two entirely different systems.

A classic example is
the exchange of data between centralized and distributed CPUs. The gateway
needs to read the network address, reconfigure the packet protocols from IPX to
SNA, translate the operating software and in most cases, completely rewrite the
data alphabet.

Communication
Transmission Devices & Techniques

Communication devices
provide basic data communications functions. They transmit, translate and
transform data into signals that can be transmitted into the physical world.
The most common communication devices are:

1.    
Modems

2.    
Multiplexers

3.    
CSU/DSU (Channel Service Unit/Digital Service Unit)

4.    
Circuit Switching

5.    
Message Switching

6.    
Packet Switching

1. Modems

Modems are necessary
because computers and transmission media speak completely different languages.
Computers are digital (0s and 1s) and most transmission media are analog (UTP
or radio waves). For digital computers to communicate over analog media, they
need translation device.

That’s where a modem
comes in. The modem actually performs two distinct tasks: modulation and
demodulation. During modulation, the modem translated digital information into
an analog waveform. Next, the data travels along the medium to the destination
device. There, the analog wave is demodulated into digital 0s and 1s.

Modem speeds are
typically measured in baud (or loosely but per second). Today’s faster modems
can support speed up to 96,000 bps (96kbps).

2. Multiplexers

Multiplexers allow
you to send multiple signals across the single transmission medium.
Multiplexing (or mixing) refers to the process of funneling multiple data
connections into one circuit for transport across a single medium. An excellent
example can be found in TV cabling.

If you think about it
for a moment, you have 100 or more channels arriving through one piece of
coaxial cabling. Your cable box or VCR demultiplexes the signal and separates
the channels.

3. CSU/DSU

Modems allows you to
connect digital computers to analog transmission media. Most WAN providers
require that you lease the CSU/DSU device from them. This ensures system-wide
continuity and safety.

4. Circuit Switching

Circuit switching
connects the sender and receiver by a single path during the conversion.
Telephone switching equipment, for example, uses address number (in the form of
a country area, office, trunk codes, etc) to establish a path that connects the
sender’s telephone to the receiver’s telephone.

In circuit switching,
a complete path must exist before communications can take place. The computer
which is initiating the data transfer establishes the dedicated path and
ensures reliable packet delivery.

5. Message Switching

Message switching
does not establish a dedicated path between two stations. Instead, conversions
are divided into messages. Each message is packaged with its own destination
address and then transmitted from the router through the internetwork.

Each router receives
the messages, stories briefly and then transmits it to the next device. This
type of network is sometimes called a store-and-forward WANs.

6. Packet Switching

Packet switching
combines the advantage of both circuit and message switching. Packet switching
breaks the datagrams into small parts called packets. Each packet is
constructed with source and destination address that allow it to work its way
through internetwork and find its destination.

Because packets have
strictly defined maximum lengths, they can be stored in RAM instead of disk,
which makes store-and-forwarding faster and easier. In general packet switching
routers use one of the two different Strategies:

1.    
Datagram packet switching

2.    
Virtual packet switching

Advantages &
Disadvantages of Network

The following are the
distinct notes in favor of networking:

1.    
The computers, staff and information can be well managed.

2.    
A network provides the means to exchange data among the
computers and to make programs and data available to the people.

3.    
It permits the sharing of the resources of the machine.

4.    
Networking also provides the function of back-up. The backup
capability is especially useful in systems like air traffic control where the
counterpart can assume its functions and workload.

5.    
Networking provides a flexible networking environment. Employees
can work at home by using through networks tied through networks into the
computer at the office.

Open System Interconnection
(OSI) Model

INTRODUCTION

OSI model has become
the primary architecture for the computer network. It was originated in 1982
and has become the most efficient ISO standard network model. The OSI model
describes how information moves through a network medium. There are seven
layers of OSI model. Each model is specific in its task.

As the message travels up the OSI Model on
the other side, the headers and footers are stripped away, finally revealing
the message at the top. To differentiate the evolution of the message as it
moves along the model, different names are used to identify the message. The
following names are commonly used.

LAYER	DATA NAME
Application Layer	Message
Presentation Layer	Message
Session Layer	Message
Transport Layer	Datagrams or segments
Network Layer	Packets or datagrams
Data Link Layer	Frames
Physical Layer	Bits

Physical
Layer

The physical layer
handles low-level rules for transmitting bits. This layer encodes or decodes
bits and sends or receives the stream of data. This is the pavement of the
information superhighway. The physical layer defines:

·        
Electrical properties

·        
Transmission media

·        
Transmission devices

·        
Physical topology

·        
Data signaling

·        
Data synchronization

·        
Data bandwidth

Electrical properties
define the rules how the message travels over transmission media, which then
create bound or unbound pathways for transmission of bits. Again, these media
compose the concrete of the information superhighway. Transmission devices
provide midpoints and functionality to the transmission media. They are the raw
implementation tools. Without translation devices, bits would scatter aimlessly
through space. Physical topologies provide the form and function. They are
defined by connection types and geographical arrangements of networking nodes.
They often determine which medium is best certain circumstances.

An example includes
bus, star, ring and cellular topologies. Data signaling encompasses coding and
timing rules for digital and analog communications. These rules govern how the
bits are received at the other side. Data synchronization defines one of two
states for the electronic transmission message: asynchronous or synchronous.

Finally, data
bandwidth is the capacity of transmission media, which determines the
communications methodology and transmission speed. Most LANs are baseband
communications, whereas WANs operate in the broadband arena.

Data Link
Layer

The Data Link Layer
organizes physical bits into logical groups called frames. These are contiguous
series of bits grouped together as a unit of data. The Data Link Layer also
detects and sometimes corrects errors. It also controls data flow and
identifies computers on the network through physical addressing.

To accomplish all
this, the Data Link Layer is organized into two sub-layers:

·        
Media Access Control (MAC)

·        
Logical Link Layer (LLC)

Media
Access Control (MAC):

These protocols are
the rules of the road. Your data signals must use one of the three different
protocols to gain access to the media. They are:

·        
Contention

·        
Token passing and

·        
Polling

Physical addressing
distinguishes one location from another. The data frames need address so that
they know where to go and what to do. Addressing works on the hierarchy
structure. The data frames need three addresses. They are:

·        
Physical address

·        
Logical address

·        
Service address

Logical
Link Control (LLC):

This layer consists
of the Functions:

Frame
synchronization: It forces senders and receivers to agree on the boundaries of
data frames. This takes place as follows:

1.    
Asynchronous – sends each frame separately

2.    
Synchronous – sends on the basis of synchronized clocks

3.    
Isochronous – incorporates advantages of above both

Flow Control: They are typically
made up of devices with different transmission speeds like storage and
processing capabilities. In diverse world, we need to develop rules that
protect slower devices but don’t hinder faster ones. These rules are called
flow control. Data Link flow control regulates how much data can be sent over
the transmission media in a specified period of time. That is achieved by:

·        
Window flow control

·        
Guaranteed rate flow control

Error Checking: It uses a complex
calculation called the Cyclical Redundancy Check (CRC). CRC ensures the
integrity of the packet by calculating a 16-32 bit number based on the contents
of the packet.

Network
Layer

Network Layer is
concerned primarily with moving data from point A to point B. In general,
network layer has four functions:

·        
Logical Addressing

·        
Switching

·        
Routing

·        
Network Control

Logical
Addressing

Logical addressing
helps to determine a logical path during internetwork communications. The
network relies on three types of addresses:

1.    
Physical address

2.    
Logical address

3.    
Service address

Switching

Most large
internetworks consist of multiple physical paths between the sender and receiver.
Intermediate routers are responsible for moving the packets across these
diverse links. Switching makes this possible.

·        
Circuit Switching: It connects the sender and receiver by
a single path during the conversation. Telephone switching equipment, for
example, uses address number (in the form of the country area, office, truck
codes, etc) to establish a path that connects the sender’s telephone to the
receiver’s telephone.

·        
Message Switching: It does not establish a dedicated path
between two stations. Instead , a conversation is divided into messages. Each
message is packaged with its own destination address and then transmitted from
router to router through the internetwork.

·        
Packet Switching: It combines the advantage of both circuit and
message switching. Packet switching breaks the datagrams into small parts
called packets. Each packet is constructed with source and destination address
that allow it to work its way through internetwork and find its destination.

Routing

Routing intermediate
network devices are used to make intelligent decisions about how the message
should travel from point A to point B. Routing involves two simple steps:

1.    
Route discovery and

2.    
Route selection

Network
Control

Network control
administers internetworking and routing. These tasks will ensure that we
maintain a steady speed and rejoin with all the rest of the datagrams at the
other end. Network layer administrative task include:

·        
Network flow control

·        
Network sequencing

·        
Network error checking

Transport
Layer

It organizes packets
into segments and delivers them reliably to upper- layer services. If packets
are not delivered to the destination correctly by the network layer, transport
functions come to rescue. This layer can initiate retransmissions and inform
the upper layers that we are trying again.

The OSI Transport
Layer consists of the following components:

1.    
Service addressing

2.    
Segmentation

3.    
Transport control

Service addressing: It provides a
doorway to upper-layer services. Remember, this is the last point of
connectivity before we get to the actual network services. Each message is
destined for a particular service. The transport layer uses connection Ids,
ports and sockets to make sure that we find the right one.

Segmentation: It is the
housekeeping task that package messages into acceptable sizes. Many upper-layer
services need specified-sized messages and segmentation does the trick.

Transport control: It includes error
checking and flows control. Error checking is the Transport Layer’s main
function. At this point, we are concerned mostly with the sage arrival of the
network messages. We don’t care what they look like. That’s the Data Link
Layer’s job.

Session
Layer

The Session Layer
opens a dialogue between the sender and receiver. It also monitors
conversations and makes sure that everyone is getting along.

The session Layer
does all this by using three simple steps:

1.    
Connection establishment

2.    
Data transfer and

3.    
Connection release

Connection
establishment: It initiates the dialogue between two systems. It uses
networking protocols to find a dialogue path and communications media to send
messages back and forth (data transfer).

Data Transfer: During data
transfer,the session layer manages the dialogue and ensures reliable
conversations through simplex, half duplex or full duplex transmissions.
Finally, connection release ends the dialogue and closes the connection.
Incidentally, connection release can be either planned or accidental.

Connection release: Finally, connection
release ends the dialogue and closes the connection. Incidentally, connection
release can be either planned or accidental.

Presentation
Layer

The Presentation
Layer is the OSI translator. It transforms the upper-layer message into a
mutually agreed-upon format. At this point, the dialogue is open and data
transfer has begun, but before the Application Layer can read the message, the
Presentation Layer must take care of two things.

1.    
Translation

2.    
Encryption

Translation: It is the
Presentation Layer’s main task. It is necessary when two system speaking
different language try to communicate most of the time. The translation can
happen in the variety of ways like Bit order, byte order, the character code
and file syntax.

Encryption:It is a secondary
task for the Presentation Layer. It is necessary for sensitive data and
operating system security, such as default authentication. This level of
encryption and decryption supports both public and private keys.

Application
Layer

The Application Layer
sits at the very top of the OSI Model. It is the ultimate goal of networking
i.e. service providing. The Application Layer does the services task as:

1.    
Service Advertisement and

2.    
Service Availability

The Application Layer
uses special networking protocols to provide file, print, message, application
and database services. Service advertisement lets other systems and users know
what services are available. Providers use active or passive techniques to
define the scope of their network services.

Service Availability is the next step. Once a
service has been advertised, it must be made available. Service availability
can be accomplished in several ways including OS Call Interception, remote
operation, and collaborative computing.