Exploiting arrays with multiple invariances using MUSIC and MODE

This paper describes several new techniques for direction of arrival (DOA) estimation using arrays composed of multiple translated and uncalibrated su barrays. The new algorithms can be thought of as generalizations of the MUS IC, Root-MUSIC, and MODE techniques originally developed for fully calibrat ed arrays. The advantage of these new approaches is that the DOAs can be es timated using either a simple one-dimensional (I-D) search or by rooting a polynomial, as opposed to the multidimensional search required by multiple invariance (MI)-ESPRIT. When it can be applied, the proposed MI-MODE algori thm shares the statistical optimality of MI-ESPRIT. While MI-MUSIC and Root -MI-MUSIC are only optimal for uncorrelated sources, they perform better th an a single invariance implementation of ESPRIT and are thus better suited for finding the initial conditions required by the MI-ESPRIT search.