Robust adaptive beamforming via subspace for interference covariance matrix reconstruction

Adaptive beamforming may cause performance degradation when model mismatch errors exist. In this paper, we have developed subspace methods for robust adaptive beamforming (RAB). The two proposed methods utilize the orthogonality of subspace to reconstruct the interference covariance matrix (ICM). Above all, the steering vector (SV) of desired signal in proposed methods is estimated from the desired signal covariance matrix, where desired signal covariance matrix is reconstructed based on the modified Capon spatial power spectrum estimator and it contains the less useless components. In the first proposed method, the ICM is reconstructed from the projected snapshots. Through projection, the desired signal in snapshots is eliminated and the interference components is retained to reconstruct the ICM. Based on the proposed-1 method, the peaks of spatial power spectrum corresponding to the reconstructed ICM indicate the nominal interference SVs which are used in the second proposed method. In the second proposed method, we reconstruct the ICM by each interference SV and corresponding power, and all interference SVs are exploited by solving a quadratic convex problem, where it depends on the orthogonality between interference SVs and interference subspace. The proposed-2 method is the extension of proposed-1 and it can achieve the better performance. Simulation results demonstrate that the two proposed methods are robust against types of mismatch to achieve well performance. (C) 2019 Elsevier B.V. All rights reserved.