DOA estimation of distributed source based on Jacobi-Anger expansion

A novel signal model of distributed source is presented based on Jacobi-Anger expansion. Firstly, the coefficients of expansion are estimated as the solution of a set of linear equations using sample covariance. Secondly, the angular power density (APD) is obtained from the estimated coefficients of expansion. Finally, the mean direction of arrival (DOA) is obtained through 1-dimensional peak searching of the estimated APD. Unlike the existed algorithms, the limitation to the width of the angular spread is avoided. Numerical examples are included to show the improved accuracy of DOA estimation in the case of widely distributed sources.