DOA Estimation Based on Maximum Mixture Correntropy Algorithm for Impulsive and Input Noise

Abstract

Maximum correntropy criterion (MCC) is widely used and plays a good performance on impulsive noise. However, it is limited by the kernel width. In this brief, to further improve the performance with single kernel function of MCC, a maximum mixture correntropy criterion is used for direction of arrival (DOA) estimation under impulsive noise, which is derived by the convex combination of two mixture correntropy (MC). Furthermore, in order to solve the problem of input noise, a new bias-compensation maximum mixture correntropy (BCMMC) is proposed, which the compensation term is derived in complex domain according to the bias-compensation (BC) criterion. Finally, simulation results show that the proposed algorithm obtain better estimated performance in impulsive noise environment than other adaptive algorithms.