Parallel architectures for multirate superresolution spectrum analyzers

The authors develop parallel architectures for implementing matrix-based superresolution spectral estimation algorithms for situations that required high levels of resolution commensurate with large coherent apertures and large sample orders. The featured architectures couple matrix-based superresolution algorithms together with a front-end multirate decimation processor. This procedure creates parallel pseudo-apertures corresponding to different subbands of the temporal (or spatial) frequency spectrum. The overall superresolution of the large aperture is maintained. Simulations applying the large array architectures to the Tufts-Kumaresan reduced-rank modified covariance algorithm and the linear-minimum-free-energy/regularized form of the modified covariance algorithm are given for 1024-element coherent apertures