Efficient, parallel adaptive eigenbased techniques for direction of arrival estimation and tracking

Abstract

Eigenspace decomposition is used in source location estimation, high-resolution frequency estimation and beamforming problems. In each case, either the eigenvalue decomposition (EVD) of a covariance matrix or the singular value decomposition (SVD) of a data matrix is required. The authors address the problem of recursive updating the EVD of a covariance matrix given the EVD of the previous matrix. This recursive algorithm is developed for multiple target angle tracking in a nonstationary environment. Simulation results include the numerical performance of this algorithm as well as its performance in high-resolution angle tracking.<>