Analysis of a joint space-time DOA/FOA estimator using MUSIC

Abstract

We examine a high-resolution signal estimation algorithm which can simultaneously estimate the spatial direction /spl theta//sub i/ and temporal frequency f/sub i/ of each source by extending the MUSIC algorithm to use space-time information obtained from one spatial sub-array and one temporal sub-array. Since the dimension of the autocorrelation matrix is flexible (due to the use of the time dimension) and can be chosen larger than the number of sources regardless of the number of array elements, the algorithm can perform joint estimation with only two array elements. Mathematical analysis is used to determine the complexity and compute the Cramer-Rao lower bound (CRLB). Computer simulation results are presented to demonstrate the algorithm's performance.