Robust adaptive beamforming based on interference-plus-noise covariance matrix reconstruction method

Abstract

Aiming at the problem of look direction error in the desired signal, a robust adaptive beamforming method based on covariance matrix reconstruction is proposed. The sparse Bayesian learning (SBL) is performed to acquire the true signal direction and the spatial spectrum simultaneously. Furthermore, the SBL spatial spectrum is used to reconstruct the interference-plus-noise covariance matrix. Compared with other reconstruction algorithms, this approach can realize the position estimation without any optimization procedures. Theoretical analysis, simulation results, and water pool experiments demonstrate the effectiveness and robustness of the proposed algorithm.