A DOA Estimation Algorithm Suitable to Small Array of Vector Hydrophones

Abstract

ESPRIT algorithm has a lower resolution when the array size is small. An improved method of ESPRIT based on unitary transformation is proposed and is applied in the vector data processing. The output data of the vector sensor array and its conjugate part form a matrix which is centro-Hermitian. Then unitary transformation is used and a covariance matrix with real elements can be obtained subsequently. The following operations, such as Singular Value Decomposition (SVD) and Eigen Value Decomposition (EVD) can be operated in the real domain. So the computation complexity becomes much smaller. As the conjugate data of the vector sensor array output is used during the DOA (Direction-Of-Arrival) estimation process, the proposed method can get a result with more accuracy and higher resolution, which is still holding under small array size. At last, several simulations are given which demonstrate the effectiveness of the algorithm.