"The greatest problem in communication is the illusion that it has been accomplished." - George Bernard Shaw

THINK about these HOME LEARN about me TASTE these links SEE my pictures SIGN / VIEW my guestbook     Last Update : Aug 16, 2003  My Research Work  Bluetooth  MSP430  Free Space Optics  Electrical Engg.  Ad-Hoc Networks  Ericsson BT Modules  Linear Prediction  Multi-user Detection    Wireless Channel Model

COMMUNICATION SYSTEMS                                                                                                           Shannon's 1948 paper

Communicating what we want to is the basis of society and doing this efficiently over inaudible distances is what communication engineering is all about!

A simple communication system is shown in the figure. It consists of various blocks.

• Source is the source of the 'information' that is to be transmitted across to the other side. It could be anything that generates something probabilistically. The fact that the source is probabilistic in nature is a basic assumption of communication engineering as otherwise we will know what the source is going to generate beforehand and hence there will be no need of communicating!
• Channel Matching or Encoding is used to convert the source generated data (which can be in any form) into entities (called signals) which are supported by the channel. Thus, we need to have a efficient way of mapping the source output to the channel input such that we can reproduce the transmitted signal at the receiving end with the least distortion. This is known as the Signal Design Problem.
• The Channel is the core of the communication system and the design of the system is completely based on the type of channel that is under consideration. Common examples of channels are the wireless channael, copper cables and fiber optic links. Thus, in order that the system performs efficiently, one has to make sure that the channel is well understood and that it is characterised by the right equations. Using the channel properties, one can then decide how to map the source output to the channel input. This area is sometimes referred to as Channel Modelling.
• Channel Decoding is in some sense the inverse of the Channel Encoding with a small difference. Since the channel in between will distort the signal, thus, the received signal may not be the one that was transmitted. This may lead to indecision in the reverse mapping and thus causing error. Thus efficient receiver design techniques are essential for minimising cases when we have errors in the received signal.
• The Sink is the actual receiver of the 'information'. It may be you when you listen to your friend on the phone, or it may be your computer when you download something from the internet.

Note: The term 'information' has a very precise meaning in communication theory and there is a whole branch dealing with this area called INFORMATION THEORY. e.g. When you see the weather report and note that Sahara Desert is hot and dry, you donot gain as much information as when the news says that there is heavy rain in that region. This is due to the fact that there are very little chances of heavy rains in the Sahara Desert and hence you get more information here than the case that Sahara Desert is hot and dry -- something which you already knew and expected to happen.

No talk about communication systems is complete without the mention of Claude E. Shanon -- the father of Modern Communication Theory. Here is the classic paper by the great man published in 1948. Come to think of it, this single paper contains almost everything that I know till date about communication systems and much more!!!