A long line array broadband adaptive beamforming with post-filtering and incoherent noise reduction

Aiming at the problem of passive broadband detection for long line array, a subarray adaptive beamforming algorithm with broadband post filtering and incoherent noise reduction based on the spectral characteristics of signals is proposed. Firstly, the full array is divided by non-overlapping and non-uniform configuration, and the subarray cross-spectral density matrix (CSDM) is obtained. The minimum eigenvalue of CSDM is used as the incoherent noise power estimation, and we exploit the diagonal reduction of subarray CSDM to alleviate noise energy. On the other hand, the eigenvectors of CSDM are projected onto the subarray steering vector, the eigenvalue corresponding to the maximum correlation coefficient is used as the signal plus noise power estimation in this direction. Finally, the frequency domain power estimation of signal and noise are proposed to construct the frequency and beam two dimensional post filter. The output of each sub-band adaptive beamformer is post-filtered to obtain the broadband result. The simulation and sea experiment show that the proposed algorithm can effectively improve the broadband detection and multi-target resolution capability of adaptive beamformer, and effective improve the output signal-to noise ratio(SNR) of target.