DoA estimation using compressive beamforming in shallow ocean using acoustic vector sensors

In this paper, we have broadened the scope of Direction of Arrival (DoA) estimation using compressive sampling in free space to a shallow ocean scenario, considering the complexities that arise due to the formation of modes in shallow ocean, along with the use of acoustic vector sensor array. We use compressive sampling of the array elements, followed by compressive beamforming, which helps in greatly reducing the complexity of the front-end circuitry and the size of array correlation matrix that is used in DoA estimation in MUSIC/MVDR algorithms. Also, in shallow ocean, the plane-wave DoA estimators generally yield biased estimates, due to the multimode nature of acoustic propagation. To solve this problem, the normal mode based MUSIC (NM-MUSIC) method is modified to be used with AVS array and compressive beamforming. In this paper, we propose a method that takes a very small set of informative measurements from an AVS array deployed in a shallow ocean, that still allow us to estimate DoA´s. Simulation results for the Pekeris model of the ocean show that resolution obtained using the compressed AVS array is similar to that obtained using a conventional AVS array in which larger number of sensors with higher snapshots are used.