LINEAR SYSTEM AND RANDOM NOISE PROCESSES

Early investigators in the field of communications first realized that the presence of unwanted random noise was an important factor following the discovery that the maximum gain of an amplifier was limited by the discrete nature of currents in electron tubes. Called shot effect, this was first explained by W. Schottky and later by many other investigators. Much research on this problem during the second and third decades of the twentieth century finally led to the rigorous formulation of the phenomenon by B. J. Thompson and others in 1940. Concurrently, the problem of spontaneous thermal noise effects in conductors was studied and formulated. By 1940, the situation was developed to an extent that the application of mathematical statistics to explain and solve broader noise problems in systems was inevitable. About this time, the basic contributions of N. Wiener led to an understanding of the optimum linear filtration of signals imbedded in random noise. His work influenced the entire course of development of theory on the optimization of filters designed to abstract a signal out of its noisy environment.