WARD: A Weighted Array Data Scheme for Subspace Processing in Impulsive Noise

This paper is concerned with robust array signal processing in impulsive noise environments. A simple weighting signal is defined to weight all sensor data in a snapshot-by-snapshot way, so that the resulting array data have the desired statistical characteristics used in the subspace-based direction-of-arrival (DOA) estimation techniques. Then any traditional subspace-based technique can be used for DOA estimation. In working example, the MUSIC algorithm is employed and a weighted array data-based MUSIC (WARD-MUSIC) algorithm is detailed. Simulations show the performance advantages of the WARD processing over other related methods in Gaussian mixture noise and symmetric alpha-stable noise