Kernel Minimum Error Entropy based Estimator for MIMO Radar in Non-Gaussian Clutter

In this paper, a kernel minimum error entropy (KMEE) based estimator is proposed for the estimation of multiple targets’ direction departure (DOD), arrival (DOA), and Doppler shift with input output radar in presence non-Gaussian clutter. Most existing approaches are on optimization complex cost function which often leads to sub-optimum solution. Therefore, accurate DOD, DOA shift, an efficient, adaptive filter (KAF) approach proposed. The utilizes (MEE) criterion minimizes function. MEE, being information-theoretic criterion, optimizes higher-order statistics thus makes robust against effects outliers like KMEE without any sparsification suffers from linear increase computational complexity. Thus, subsequently, complexity reduced by incorporation novelty (NC) technique, resulting called KMEE-NC. performance KMEE-NC compared recently introduced sparse estimators maximum correntropy mean square criterion. Additionally, also other conventional estimators. Further, assessing accuracy estimator, modified Cramer-Rao lower bound derived using Fisher information matrix.