A robust beamforming algorithm based on optimal reconstruction of covariance matrix

DOI:	10.11809/bqzbgcxb2023.09.041
Keywords:	adaptive beamforming; optimal reconstruction of covariance matrix; steering vector estimation; uncertainty set constraint; iterative solution
Abstract:	Focusing on the problem that serious performance degradation of general adaptive beamformers in the presence of model mismatch error, a robust beamforming algorithm based on optimal reconstruction of interference plus noise covariance matrix is proposed in this paper. Firstly, the expected signal components in the sample covariance matrix are eliminated by sparse reconstruction method to estimate the interference plus noise covariance matrix. Then, based on the knowledge of subspace expansion, the interference plus noise covariance matrix is optimized by establishing steering vector uncertainty set constraints. After that, a convex optimization model of target steering vector is established in order to maximize array output power and correct the target steering vector. Finally, the model is solved by cyclic iteration method and the optimal weight is obtained. Theoretical analysis and simulation results demonstrate the robustness of the proposed algorithm under target arrival direction error and array position error.
Published:	2023-09-28
PDF HTML