DOA Estimation Using Sparse Vector Sensor Arrays

Abstract—Acoustic vector sensors (AVSs) have several ad-vantages over traditional scalar sensors as they measure the acoustic velocity in addition to the acoustic pressure, which is measured by traditional scalar sensors. Direction-of-arrival (DOA) estimation using sparse scalar-sensor arrays that provide O(M2) degrees of freedom using M sensors has been studied extensively. In this work, these concepts are extended to AVS arrays. More specifically, a spatial-velocity based smoothing technique is proposed to utilize the O(M2) degrees of freedom. We also show that the number of sources whose DOAs can be identified is significantly greater than the number of physical AVSs. We verify the effectiveness of the proposed approach through extensive simulations. I