ABSTRACT
This chapter provides a sharp deviation from the central theme of independent and identically distributed clutter returns in the clutter range profile (CRP) A formulation of a general OS-based decision rule, which was then specialised to the case of clutter modelled by K. V. Mardia's multivariate Pareto model. Real radar clutter returns will often be correlated spatially and hence it is important to account for this in the design of decision rules. This will result in an inevitable complication in the derivation of sliding window constant false alarm rate (CFAR) detectors. The developments above discussed have shown that the CFAR property can only be achieved for the Pareto scale parameter, for the case of correlated Pareto distributed clutter modelled by the Mardia multivariate model. An assumption made throughout was a priori knowledge of the Pareto shape parameter. The decision rules examined are CFAR with respect to the Pareto scale parameter.
