LINEAR FILTERING OF OUTPUT DATA OF A SINGLE-POSITION SECURITY SYSTEM

Abstract

Obtaining an estimation of the filtering quality of a single-position security system is described in the paper. A general description of problems of estimation of filtering quality in the system is presented in the introduction. A geometric construction of a single-position security system, its mathematical description and formulas of Markov sequence that describes the motion of an object under observation are given. Descriptions of the processes of estimation with the use of α-β filtering, Wiener filtering, Kalman filtering for the system are presented. A definition of estimation describes criteria for the quality of filtering process, parameters for modeling the system and the results of the modeling. The suggested criteria for the quality of filtering process include average, variance and standard deviation of filtering errors and interval between the true and filtering values. Initial values for the modeling and the results of the modeling are presented. In conclusion, the filtering improves with increasing spectral width of probing signal and signal/noise ratio. However, Kalman filtering is most effective.