On the eigenstructure of the signal-only tempo-spatial covariance matrix of broad-band sources using a circular array

The eigenstructure of the signal-only part of the data covariance matrix for broadband sources is considered. It is well known that, unlike the narrowband case, all eigenstructure parameters (the matrix rank, eigenvalues, and eigenvectors) depend on source spectral and spatial parameters, as well as on the relationship between sources. It is proved that when a two-dimensional equally spaced circular array with flat-spectrum broadband sources is used, the eigenstructure of the signal-only tempo-spatial covariance matrix (TSCM) has properties similar to those for the narrowband case. Most importantly, the effective rank of the TSCM depends only on the number of sources. Consequently, the authors suggest an alternative approach for resolving frequency/location ambiguity in the broadband case that may lead to an effective eigenstructure-based parameter estimation algorithm of the MUSIC type