Most Wide Area Network standards have been developed for use by public-carrier data networks. A public data network is a network established and operated by a national network administration authority specifically for the transmission of data. A fundamental requirement for a PDN is that it should facilitate the interworking of equipment from different manufacturers, which in turn requires agreed standards for access to and use of these networks. After much discussion and experimentation at national and international level, a set of internationally agreed standards have been accepted - these X-series and I-series recommendations include standards for user data signaling rates and user interfaces with such networks.

There are two main types of PDN - circuit switched and packet switched - each with its own standards.

Circuit Switching

Each connection through a circuit switched network results in a physical communication channel being set up through the network from the calling to the called subscriber equipment. The circuit may pass through a number of telephone exchanges in different parts of the country. At each stage a switch is physically closed until a link is made between the two ends. The physical circuit will remain for the duration of the call and at this time it is said to be BUSY.

In some instances there may be significant traffic over a line that in fact a permanent line is required. This permanent physical link can be achieved by means of a dedicated line, of two types - a private line or a leased line.

If the distances involved are very short, for example in a single factory, a private line may be implemented. This would be installed by, and be the property of the company concerned. However, if a public road, or any land not belonging to the company, is to be crossed, there is a legal requirement to lease a line from an authorised common carrier, for example British Telecom.

On the other hand, a leased line could be useful for a company with its headquarters in one part of the country and a branch office in another part. Traditionally the supplier of leased lines is British Telecom, however, in the last few years they have faced competition from Mercury Communications. Although the leased lines would be used exclusively by the company concerned, they would remain the property of the common carrier.

Advantages

Disadvantages

Packet Switching

Packet switching can be defined as the routing of data in discrete quantities called packets, each of a controlled format and with a maximum size. This type of system was originally designed to provide reliable transmission through a communications network where switches or lines may suddenly become unavailable or alternative routes may have to be chosen. A packet switching network consists of a number of computer-based switching nodes connected via high-speed communications lines, as shown below:

A packet consists of a "header" section which contains information such as the network address of the destination terminal, a data section containing the information to be transferred and a "tail" section containing checking information. A stream of data is therefore split up into a number of packets when these techniques are employed. Packets may vary in length from a few bytes up to several thousand. Once a message has been transmitted, the packets which comprise it, may take different routes through the communications system and arrive at the Receiver at different times. In this scenario it is up to the Receiver to reassemble the packets into the correct order, to obtain the complete message.

In order to allow network users to exchange packets, there must be a standard interface protocol between the network and the computers and terminals connected to it. CCITT recommendation X.25 is now established as the interface standard between terminals and packet switching networks. X.25 is compatible with the bottom three layers of the < a href="iso.htm">ISO Reference Model. To get away from proprietary network techniques many private and public users are attempting to standardise on X.25 packet switching for data networks.

Some systems operate only at the packet level and as stated earlier it is possible for packets to arrive at the Receiver in a different order from that in which they left the Sender these systems are known as datagram systems. Moreover, some systems employ a routing technique known as adaptive routing, which the system will attempt to work out the most efficient path for any packet to take through the network.

On the other hand, some systems operate at the message level and thus maintain the packets making up a message in the correct sequence - virtual circuit systems. To the receiver they function in an identical fashion to circuit switched systems, however, in this case the network must cater for the retransmission of lost or damaged packets and the elimination of duplicates.

Advantages

Disadvantages

Message Switching

One last form of switching is known as Message Switching. Data terminals and computers are interconnected by using circuit or packet switching as previously discussed. However, administrative message traffic is more efficiently handled using the "store-and-forward" technique. Unlike circuit switching no "call" is established in a message switching system. A message is stored and then forwarded as soon as the destination is free to receive traffic, either other leased lines or over the public dialled network using fully automatic dialling.

Message switching systems have the following qualities: