DOA estimation method for incoherently distributed sources based on spatial-temporal generalized ESPRIT

This paper proposes a spatial-temporal generalized ESPRIT parameter estimation method to solve the estimation problem of the central angles and spread angles of incoherently distributed sources. To begin with, a virtual array covariance matrix is built by conducting time-delay autocorrelation operation on the original received data. The central angles are then estimated with the generalized shift invariance property inside the virtual array manifold. Next, the spread angles are obtained with resorting to the estimated central angles and the central moments of the angular distribution function. Compared with the existing algorithm, the proposed algorithm does not require the signal to have non-circular characteristics, and has higher parameter estimation accuracy and can distinguish more sources. In addition, the proposed method is robust to types of spread angles, as verified by a series of simulation results.