Robust Adaptive Beamforming Using a RCB Principle Modified Subspace Method

Abstract

In this paper, we propose a novel robust adaptive beamforming algorithm by constructing subspaces through the robust Capon beamformer (RCB) principle. The proposed algorithm consists of two steps: reconstructing the interference-plus-noise covariance matrix (IPNCM) and estimating the steering vector of the desired signal. In the first step, we improve the constructed interference-noise subspace by the RCB principle and obtain the signal-interference subspace by the sample covari-ance matrix. Then the steering vectors of the interference signals are estimated by the projection of these two subspaces, and the IPNCM is reconstructed by the combination of these steering vectors. In the second step, we use the RCB principle to improve the estimation of the signal subspace and then take its principal component as the estimation of the steering vector for the desired signal. The adoption of the RCB principle improves the accuracy of the reconstructed IPNCM and the estimated steering vector of the desired signal; hence this strategy improves the overall performance of the beamformer. The simulation results demonstrated that the proposed beamformer outperforms the other conventional beamformers in output signal-to-interference-plus-noise ratio under various mismatched models.