The Shannon&Weaver Model

The most commonly accepted and spread communication model is that proposed by electrical engineer Claude Shannon in 1949 and then interpreted by Warren Weaver. It must be noted that particularly Shannon's model was intended to be an information transmission model, mostly applicable for engineering purposes. Nevertheless, Weaver's interpretation, as well as further later uses converted it in a general usage communication model.

Shannon and Weaver introduced the linear model represented in figure 5.16. The goal of any communication system is to transmit a data source to a particular destination. The transmission is carried out through a physical channel that is bound to influence the message, a non-ideal transmission channel can indeed be considered a noise source. The engine in charge of preparing the data source and sending it through the channel is called the transmitter. This transmitter is in charge of different process but the most important is that of encoding the message so it can be transmitted more efficiently and it can become more robust to the effect of the channel. The receiver is in charge of obtaining the message from the channel, decoding it, removing noise as much as possible and deliver it to the final destination.

Figure 5.16: Shannon & Weaver's classical information transmission metamodel

In this metamodel, information is thought of as the opposite to entropy. Meaning is unimportant from that mathematical point of view. According to Shannon and Weaver ``the semantic aspects of communication are irrelevant to the engineering problem'' [Shannon and Weaver, 1949]. As Eric Scheirer points out this is not an engineering assumption but rather a philosophical one [Scheirer, 2001].

In a similar way, in S&W's model noise is anything added by the channel and not intended by the source. The way to overcome the unwanted effects of noise is by adding redundancy. As a matter of fact the S&W theory considers that communication is the science of maintaining and optimal balance between predictability and uncertainty by adding or removing redundancy.