Masked beamforming in frequency domain for hybrid acoustic vector sensor array

Abstract

A masked beamforming algorithm for a hybrid acoustic vector sensor array (AVSA) is presented. In the general literature, a full-rank AVSA, comprised entirely of an acoustic vector sensor (AVS), is known to be capable of left–right ambiguity elimination. However, a full-rank AVSA has disadvantages including rotation perturbation of sensor directivity, complexity of electrical loading, and high cost. To alleviate these disadvantages, a hybrid AVSA consisting of only a single AVS in an array of omni-directional sensors is reported in the patent with a limited performance in left–right ambiguity elimination. A masked beamforming algorithm is proposed to achieve good left–right ambiguity elimination performance in a hybrid AVSA by masking the beam output spectrum of the omni-directional sensor array with the beam output spectrum of an AVS in the frequency domain. Simulation results that validate the performance of the proposed method are also presented.