Geometric characterization of eigenvalues of covariance matrix for two-source array processing

For a two-source array processing scenario, the normalized eigenvalues expressions lambda /sub 1/ and lambda /sub 2/ are reduced to forms depending only on a real triplet: phase-dependent variable xi , phase-independent variable eta , and power ratio pi /sub 1// pi /sub 2/. ( xi , eta ) is confined to an isosceles-like region. This isosceles-like region is characterized along with the many-to-one mapping from the Cartesian product of the temporal and spatial correlation unit-disks onto this region. The behavior of the eigenvalues and their ratio as functions of the real triplet with respect to array processing are also discussed. A characterization is given of Speiser's (1989) eigenvalue bounds specialized to the two source scenario. >