A robust subband subarray steered minimum variance beamforming algorithm

The STeered Minimum Variance (STMV) beamforming was a fast-convergent adaptive algorithm, the weight of which are obtained using steered covariance matrix. The performance of traditional Robust STeered Minimum Variance (RSTMV) beamforming will degrade obviously for wideband signals. In order to improve the performance of RSTMV, a robust subband subarray STMV beamforming algorithm is proposed in which the frequency domain subband and spatial domain subarray techniques are combined. The weight vector can be obtained using appropriate diagonal loading amount through constraining every subband's boundary of Array Steering Vector (ASV) errors' norm. The number of subbands can be futher increased using the subarray techniques, so the performance can be obviously improved and the complexity can be effectively reduced. The theoretical analysis and simulations show the new algorithm has deepest nulling and narrowest beam width, the output Signal to Interference and Noise Ratio (SINR) was close to the theoretical value, so the performance of this algorithm was best when ASV errors exist. The results of trial data show that the output SINR of the proposed algorithm can be increased by 10 dB relative to the Conventional Beamforming (CBF) and 4 dB relative to the RSTMV beamforming when there is strong interference. At the same time, the computational complexity and resolution can be effectively improved, the power of targets can be obtained distortionlessly. © 2019 Acta Acustica.