1.5: Networks

Introduction to Networks

Network Definition:

A network is: 2 or more computing devices connected so that they can share data and resources

Standalone (non networked) Computers

If all the computers in a room are standalone, it means they are not connected to each other.

Disadvantages of Standalone Computers

They can't send messages to each other
They can't share a file easily
They can't share a hare a piece of equipment like a printer easily
Users can't work on different machines and easily access files that were saved on the first computer.

If computers were connected together in some way then they would be able to do all of the above.

- Computers that are connected together are known as a 'network'.

See above definition

LAN & WAN

Where each of the 8 bits is transmitted down a single wire connection one at a time

Parallel Data Transmission -
- - - - Using many wires to each transfer a bit at the same time
        (eg 8 wires to transfer an entire byte of 8 bits)

Serial Data Transmission -

Broadly, networks can be split into two categories, LANs and WANs.

Local Area Networks

are networks that are made up of

computers connected to each other in a geographically small area

(e.g. in a room, in a single building, or in a number of close buildings)

Wide Area Networks,

are made up of

computers connected up to each other over a wide geographical area.

The Internet is a good example of a WAN.

Companies who have offices scattered over the country may have a Wide Area Network.

Data Transmission

Networks need to transfer data between computers and other devices

'Handshaking'

There are 2 types of handshaking that could be used.

- - Hardware handshaking &
  - software handshaking

Simplex Data Transmission:

Before two devices can communicate, they must ensure they are both ready to communicate

So that one device (e.g Computer) is ready to send and the device (modem) is ready to receive,

or vice versa. How they do this is referred to as handshaking.

Data is only transmitted down the wire in same direction
When data transmission can only take place in one direction, we talk about 'simplex transmission'.
- Teletext is a classic example of simplex transmission.
- Data is broadcast by television companies at the same time as TV pictures
- People do not send back signals from their television to the aerials.
- You can use a handset to send requests for pages to your TV's special Teletext adapter and the special Teletext adapter then captures the requested Teletext pages when they get broadcast.
NB Teletext is now dead in 2010

Half Duplex Data Transmission:

Data can be transmitted down the wire in both direction,
- BUT only in one direction at a time
Simplex data transmission is one-way communication.
'Half duplex communication' is communication that can happen in both directions,
- but not at the same time
- The classic example of this is a set of walkie-talkies.
- Each handset can be used to either send or receive,
  - but cannot do both at the same time.

Duplex Data Transmission:

Data can be transmitted down the wire in both directions
Duplex data transmission describes communication that occurs
in both directions at the same time.
- The classic example of this is the telephone.
- You can both send and receive information at the same time.
- You can both talk and hear at once.

Bit Rate

As shown with parallel data transmission,

more wires connecting 2 devices enable more data (bits) to be transferred during a period of time

The amount of bits per second is know as the baud rate
this is the key measure of data transfer rates

Data Transmission Errors

Therefore, it is necessary for computers to check that data has been sent correctly and that it hasn't become corrupted.

When data is transferred from one place to another, it can become 'corrupted'.

There are 3 main ways of checking data transmission errors that we need to understand.

These are:

1. Parity checking.
2. Echo.
3. Check Sum

e.g. electrical cables and electrical devices generate a magnetic field.

This field can interfere with the electrical signals that make up transmitted data,

resulting in bits being changed from a 'one' to a 'zero' or from a 'zero' to a 'one'.

When you send a byte of information using 7-bit ASCII, you have one bit spare.

This extra bit can be used to check for errors in transmission.

By using an error-checking method known as 'parity checking', half of these errors can be detected.

Parity checking involves both devices deciding in advance whether they are going to use even parity or odd parity.

There is no advantage of one method over the other.

With even parity, the number of bits in every byte must always be even.
With odd parity, the number of bits in every byte must be odd.

An example of Parity Checking

Sending the 7-bit ASCII code 0001001.

The byte we need to send is ?0001001.

The question mark is the parity bit, which we will decide is either zero or one.

Before transmission we decided to use even parity,

The number of bits in every byte must be even.

Therefore, we must make the parity bit in this byte a zero to keep the total number of bits even.

If we had decided in advance to use odd parity for data transmissions,

we would have to set the parity bit to a one, to make the total number of bits in the byte odd.

Even Parity Checking

Parity Checking

Using even parity to send the byte 00001001
The computer receiving the byte knows that it is using even parity because it was agreed in advance.
It counts the number of bits in the byte
- If they are even it accepts the data.
If it counts an odd number of bits,
- One of the bits must have been corrupted
- It must signal to the sending computer to send the data again.

Odd Parity Checking

- Using odd parity to send the byte 10001001
- The computer receiving the byte knows that it is using odd parity because it was agreed in advance.

It counts the number of bits in the byte
- - If they are odd it accepts the data.
If it counts an even number of bits,

- One of the bits must have been corrupted
- It must signal to the sending computer to send the data again.

NB If you send 00001001 using even parity, and two of the bits get corrupted,(eg. 11001001)

There will still be an even number of bits and the data will be accepted, despite the error.

If 4 bits or 6 bits or 8 bits were corrupted in this example using even parity, there would also be no error reported.

An error will only be reported if 1, 3, 5 or 7 bits get corrupted.

The same applies to picking up errors using odd parity.

Parity checking is a useful way of picking up half of the errors created during data transmission.

Echo Checking

Another very useful way of checking if a message has been sent successfully is 'Echo Checking'.

When a message is sent, the receiving computer returns the message to the sending computer that asks,

"Is this what you sent"?

- If it is,
  - then the sending computer signals to the receiving computer that the message was sent correctly.
- If it isn't,
  - then the message is re-sent.

Echo Checking requires data to be sent twice and therefore takes longer.

Check Sum Method

Data is usually sent in blocks because that is the most efficient way of sending data.

Apart from the data, an extra byte is added, known as the 'Check Sum'.

The value of the check sum is arrived at by adding up the data bytes and discarding any carry.
When the receiving computer receives the data, it does its own check sum calculation,
and then compares it with the check sum that was sent.
- If they're the same,
  - then it accepts the data.
- If they are not the same,
  - then it can request the data to be sent again.

Page updated

Google Sites

Report abuse