A Note on the Bayesian Approach to Sliding Window Detector Development
share
Recently a Bayesian methodology has been introduced, enabling the construction of sliding window detectors with the constant false alarm rate property. The approach introduces a Bayesian predictive inference approach, where under the assumption of no target, a predictive density of the cell under test, conditioned on the clutter range profile, is produced. The probability of false alarm can then be produced by integrating this density. As a result of this, for a given clutter model, the Bayesian constant false alarm rate detector is produced. This note outlines how this approach can be extended, to allow the construction of alternative Bayesian decision rules, based upon more useful measures of the clutter level.
READ FULL TEXT