 |
Claude Shannon, another mid-western American, was born in Gaylord, Michigan in 1916. From an early age he displayed a keen interest in both engineering and mathematics, and this dual interest in theory and its practical application can be seen throughout his career. He graduated from Michigan University in 1936 with degrees in both Mathematics and Engineering and then moved on to the Massachusetts Institute of Technology to study for his Ph.D. |
Probably due to his practical engineering talents he was put in charge of the M.I.T. differential analyser, a form of mechanical computer used for solving differential equations. Programming this machine was essentially a construction project, which involved reconfiguring gear wheel assemblies and rewiring sets of electromechanical relays. However, this led to Shannon's first major work, his master's dissertation on the topic 'A Symbolic Analysis of Relay and Switching Circuits.' In this he first pointed out the relationship between switching circuits and Boole's work on the Algebra of Logic and went on to use Boolean algebra to analyse and design switching circuits. This had immediate applications in the design of telephone switching systems (i.e. Telephone Exchanges) and within a few years was proving to be one of the foundations for the design of electronic computers.
His Ph.D. studies went to the other extreme when he tackled a subject that is only now coming to the front line of research. It was a dissertation on Theoretical Genetics. In view of his contribution to switching circuits it was hardly surprising that he joined the Bell Telephone Laboratories after completing his Ph.D. He arrived at the Bell Laboratories in 1941 and remained there until 1972 . The telecommunications environment of the Bell Laboratories led to his next, and most important work.
In 1948 (soon after I started work in the G.E.C. Telecommunications Laboratories) he published a paper on 'The Mathematical Theory of Communication.' This was probably one of the most influential pieces of work published during the 20th century - in the same class as Einstein's Theory of Relativity. Shannon showed how information could be expressed as a sequence of symbols (preferably binary digits) that were amenable to mathematical manipulation and analysis. This led to measures of information content of a signal, extended Nyquist's work on sampling and digital transmission, showed how noise reduced information and from that led to techniques for error correction, and finally opened up the possibility of applying security coding (cryptography) to electronic communication. This provided theoretical foundations for the development of computers, digital transmission systems, and digital electronic telephone exchanges over the next few years.
All of these techniques have now been adopted in the field of control system design, which is, basically, a process of getting information and doing something about it.
Like many original thinkers, Shannon was something of an eccentric. In his early days in the Bell Laboratories he was noted for his habit of hiding in his office throughout the day and then emerging after dark to ride his unicycle round the corridors..
Previous page
Next page