A fast algorithm for solving periodically Toeplitz systems

This paper presents an efficient algorithm for computing the inverse of matrices which are periodically Toeplitz, representing the correlation structure of cyclostationary processes. Such matrices can not be inverted by Trench's method [5] as they are not persymmetric. This paper proposes an alternative approach based on appropriate matrix factorization and partitioning. The algorithm provides certain insight on the formation of the diagonal and subdiagonals of the inverse matrix and is implementable on a set of circularly pipelined processors.