Using Polygaussian Models to Describe Random Signals and Noises with the non-Gaussian Nature of Distribution

Polygaussian models for describing signals and noise including non-Gaussian ones in radio engineering information and measurement systems is presented. It has been shown that for a particular class of such systems used for the aerospace industry and mobile communication systems, using traditional correlation methods precludes from describing the real signal-noise situation. The polygaussian representation of random signals and noise is given. It is shown that polygaussian representations with any number of components are possible. Optimal polygaussian algorithms for receiving discrete signals are considered. The <;<;multicorrelation>> algorithm of the optimal receiver is obtained. It is pointed out that the task of distinguishing several random signals is reduced to the task of distinguishing certain polygaussian processes. The functional diagram of the polycorrelation receiver of the deterministic signal is given. Polygaussian algorithms for optimal reception of discrete signals were presented in more detail. Reception of signals with rectangular envelope under influence of pulse noise is described. Expressions for determining likelihood functionality are obtained. Functional diagrams of rectangular radio pulse receiver are given. It is shown that the obtained polycorrelation algorithm is invariant.