COVARIANCE-STRUCTURES FOR MULTIDIMENSIONAL DATA

Certain covariance constraint sets useful in a wide class of signal processing problems are described. The constraint sets define structure in the covariance matrices of naturally multidimensional data organized into column-vector form. A notation characterizing these constraint sets is established. The structures involve a hierarchy of sub-blocks within the matrix, and include block-circulant and block-Toeplitz matrices and their respective generalizations. An example describing the structured covariance for wideband data from a uniform linear array of unstructured subarrays is included.