Knowledge-aided Bayesian radar adaptive detection in heterogeneous environment: GLRT, Rao and Wald tests

Adaptive filtering is an effective method for clutter suppression and radar detection. However, the performances degrade severely if the environment is heterogeneous. To solve this problem, we resort to a Bayesian framework and design knowledge-aided detectors under partially homogeneous model assumption, which outperform their conventional counterparts in heterogeneous environment. It is also proved that the proposed Bayesian generalized likelihood ratio test (GLRT) coincides with the Bayesian Rao and Wald tests, under the assumption that the covariance matrix of the cell under test is proportional to that of the training data.