Robust Kalman-based time-varying DOA estimation of impulsive noise

Abstract

This paper presents the robust Kalman-based method which is modified with a Huber function for estimating time-varying DOA (direction of arrival) under an impulsive noise environment. The Kalman-based algorithm can estimate the proper parameter in the additive white Gaussian noise. But it is well known that the performance of the Kalman algorithm degrades seriously when the input or the estimation error contain impulsive noise (burst outliers) which can arise frequently in many practical applications. The proposed algorithm has the robustness over the impulsive noise by putting bounds to the derivative of its cost function with respect to the estimation error, using Huber function. It is known in the theory of robust estimation that if the influence function is bounded, the estimator is robust. Simulations show that the performance of the proposed algorithm is less vulnerable to the impulsive noise than the conventional Kalman algorithm.