Efficient DOA Estimation of Coherent Sources Using Unitary Projection Matrix

This paper presents a Unitary Projection Matrix (UPM) method to reduce the computation cost of Direction-Of-Arrival (DOA) estimation for far-filed narrowband coherent sources. The proposed method requires only two correlation vectors of array observation to the 1st and last sensor, in which the spatial independent additive noise is eliminated effectively. The coherency of sources is decorrelated by subarray smoothing technique. The signal subspace is then approximated using matrix algebra operation in real domain instead of eigenvalue decomposition (EVD) or the singular value decomposition (SVD) in complex domain.