Adaptive beamforming based on linearly constrained maximum correntropy learning algorithm

Abstract

The Gaussian noise profile has been demonstrated to be an inaccurate model in several antenna beamforming problems. Many available beamformers are based on second-order statistics and their efficiency degrades significantly due to impulsive noise existed in the received signal. Therefore, a demand exists for attention to address beamforming problems under nonGaussian noise environments. According to the robust performance of information theoretic learning (ITL) criteria in nonGaussian environments, we propose a linearly constrained version of maximum correntropy learning algorithm in order to solve beamforming problem in presence of nonGaussian and impulsive noises. Simulation results of the proposed adaptive beamformer are provided to illustrate its accurate and resistant performance in comparison with conventional second-order-moment-based beamformers.