Methods of Forming Classified Training Sample for Adaptation of Weight Coefficient of Automatic Interference Compensator

Abstract

The problem of analyzing various methods for the formation of the classified training sample that ensure the effective operation of the automatic compensator for active interferences with simultaneous presence of passive interference is solved in this paper. For the first time an adaptive method for the formation of the classified training sample based on the use of threshold estimation of the interchannel correlation coefficient of the combined interference is proposed. An adaptive method for the formation of the classified training sample based on the current interval estimation of the distribution of the passive component of the combined interference in range (time) is also proposed. The method provides an estimation of the coefficient of interchannel correlation with the selection of the maximum value and its utilization for the formation of weight coefficients in the following probing period. Experiment in the testing ground is conducted with a quantitative estimation of the cancellation ratio of active interference for the formation of the classification sample using spectral differences in the structure of active and passive interferences.