An Adaptive Detector in Compound Gaussian Clutter of Nonhomogenous Environments

The adaptive detection of signal embedded in compound Gaussian clutter of nonhomogeneous environments, i.e., the training samples used for adaption do not share the same covariance matrix as the vector under test is considered in this paper, and the clutter can be modeled in terms of a compound Gaussian process. With known prior and some appropriate prior distribution, based on Bayesian framework, the minimum mean square error estimation of clutter covariance matrix is proposed, and the application to the adaptive normalized matched filter test is given. The results of computer simulation are presented to illustrate that the performance of the proposed detectors is better than conventional ones, especially in the present of a small number of training data.