ALGORITHMIC METHODS OF QUASI-OPTIMAL FILTER SYNTHESIS

Abstract

The aim of the work is to create and to refine methods of algorithmic filter synthesis allowing designing quasi-optimal filters in cases when statistical characteristics of effective signal and interference – non-stationary stochastic processes – are unknown and application of classic synthesis methods, e.g. Kalman-Busy method, is impossible. Paper presents the basics of algorithmic quasi-optimal filter synthesis, the essence of which is the for-mulation of the filter synthesis task in the form of searching optimization and its solution with application of simplex search methods. Examples of algorithmic quasi-optimal filter synthesis are presented. Parameters of quasi-optimal filter transfer function found by algorithmic method correspond to theoretical parameters calculated on the basis of sta-tistical characteristics of input signals.