Unitary ESPRIT: how to exploit additional information inherent in the relational invariance structure

ESPRIT is a high-resolution signal parameter estimation technique based on the translational invariance structure of a sensor array. Previous ESPRIT algorithms do not use the fact that the operator representing the phase delays between the two subarrays is unitary. Unitary ESPRIT, however, exploits this knowledge to estimate the directions of arrival in a straightforward way yielding more accurate results. This is particularly useful in situations where the standard ESPRIT algorithm faces severe problems, such as closely spaced sources or short window lengths. The improved subspace estimates form the basis for several algorithms to separate and reconstruct superimposed signals arriving from different directions. Moreover, Unitary ESPRIT improves the performance of recently developed subspace estimation schemes, that only approximate the optimal signal subspace, but can be updated with significant computational savings.<<ETX>>