RADAR-TRACKING OBJECTS SEQUENTIAL RECOGNITION ALGORITHM, PROVIDING MINIMIZATION AVERAGE RISK OF ACCEPTED DECISIONS

Abstract

For increase of reliability of radar-tracking objects classification it is offered to use sequential Bayesian recognition algorithm, providing minimisation risk of accepted decisions on an object class and about further supervision continuation. For the information sequential specification propose to use a posteriori probability of solving statistics of the classification previous step as a priori information for a following step. Calculation results of radar-tracking objects recognition characteristics for three simulated classes to sequential algorithm and one-step Bayesian classification algorithm are produced.