Parallel structure for recursively updating the covariance matrix for real-time image processing applications

The paper presents a time optimal parallel architecture for the inversion of a special class of Range-Hermitian matrices. In particular, the paper derives recursive equations for the computation of the covariance matrices, which are a sub-class of Range-Hermitian matrices. The derived recursive equations update the covariance matrix and its inverse taking into account all the previous parameters. These equations apply for the singular and nonsingular cases. A unique feature of the architecture is the capability of online updating of the covariance matrices. The proposed architecture is capable of updating an N×N covariance matrix in N+1 cycles. It features full use of symmetry properties to speed up computations and to reduce storage requirements