Modified subspace smoothing technique for multipath direction finding

Recently a new subspace smoothing algorithm for estimation of directions of arrival (DOA) in coherent signal environment by using antenna arrays is presented. Unlike the existing forward and backward spatial smoothing algorithm, the new subspace smoothing algorithm is not restricted to uniformly linear arrays, and M — 1 DOA's of direct path and multipath components can be estimated, where M is the number of the sensors. Unfortunately the steering vectors have to be estimated several times in that algorithm, which is computationally very intensiv. In order to reduce the computational cost but keep the advantages of the new idea, a modified version of the subspace smoothing algorithm will be presented in this paper. We will show that there is equivalence between the modified subspace smoothing algorithm and the well-known forward spatial smoothing algorithm. We will also show that the array aperture is extended virtually by using the modified sub-space smoothing algorithm. Finally, simulation results will demonstrate the features of our modified algorithm.