Research On Robust Beamforming Method Of Vector Flank Array

Vector hydrophone has many advantages,such as comprehensive acquisition of sound field information,excellent noise suppression performance and so on.Therefore,the flank array composed of vector hydrophone can obtain better detection performance,but the self noise of the platform will seriously reduce the detection performance of array targets in the process of use.The spatial matrix filter can make signals in the region of interest pass while suppressing strong interference in the region of not interest,and the diagonal unloading method can eliminate the uncorrelated noise and improve the detection performance.Therefore,the research on robust beamforming method of vector flank array under platform self noise interference has important application value.First,according to the spatial distribution of the vector flank array platform self-noise in the actual situation,a simplified platform interference noise model is established,the theoretical characteristics of mechanical noise,propeller noise and flow noise are analyzed,and combined with the far-field target,a mixed signal processing model of the vector flank array under the influence of the platform’s self-noise is established.Secondly,aiming at the problem of platform noise suppression,four optimized designs of vector flank array spatial matrix filters are proposed.The simulation and comparison analysis of the influence of different parameters on the performance of the four spatial matrix filters is concluded,and it is concluded that it is suitable for vector flank arrays.Based on the optimal spatial matrix filter,the influence of different parameters on the performance of MVDR target orientation estimation based on the optimal spatial matrix filter is simulated and analyzed.The simulation results show that the optimal spatial matrix filter can ensure the suppression of platform mechanical noise and propeller noise while allowing the signal to pass through the detection area without distortion.Finally,in order to solve the problem that the flow noise and the residual platform noise after spatial filtering affect the performance of the subsequent beamforming method,a MVDR target orientation estimation method based on uniform and non-uniform diagonal unload is proposed.This method can eliminate most of the uncorrelated noise components on the diagonal of the pre-filtered covariance matrix,and improve the performance of subsequent beamforming methods.Subsequent simulations compared the pre-filtered vector MVDR direction of arrival estimation method,and the performance of the MVDR direction of arrival estimation method based on uniform and non-uniform diagonal unload.And through the simulation results,analyzed the influence of the number of array elements,the number of snapshots,the environmental noise power after pre-filtering,and the flow noise power on the performance of MVDR direction of arrival estimation based on uniform and non-uniform diagonal unload.Simulation results show that both uniform and non-uniform diagonal unload methods can improve the performance of MVDR direction of arrival estimation after pre-filtering,but when the noise components on the diagonal of the covariance matrix are inconsistent,the performance of the MVDR direction of arrival estimation method based on non-uniform diagonal unload is better than MVDR direction of arrival estimation based on uniform diagonal unload.