Studies of split spectrum processing, optimal detection, and maximum likelihood amplitude estimation using a simple clutter model

This work continues our earlier comparison of split spectrum processing (SSP) and prototype or geometrical based optimal signal processing for clutter suppression. Presented results selected from a large set of experiments on simulated data generated by a new simple clutter model show that the prototype based approach is very competitive. Experiments using both nearly Gaussian and non-Gaussian clutter are considered. The clutter model is shown to refine some recently published models and is used to support the conjecture that clutter caused by relatively large grains often can be well approximated by a Gaussian stochastic process. This work also shows that the geometrical approach can be formulated as a new SSP method with a well defined filter bank structure which yields both optimal detection and maximum likelihood amplitude estimates.