Maximum-Likelihood Estimation for Covariance Matrix in Compound-Gaussian Clutter via Autoregressive Modeling

This paper addresses the problem of speckle covariance matrix estimation for Compound-Gaussian clutter. The speckle component is modeled as a low order autoregressive (AR) process. We derive the AR coefficients conditioned Likelihood function of the secondary data and propose an iterative approach for the optimizing problem under the criteria of Maximum-Likelihood (ML). We evaluate the performance of the new method by the normalized Frobenius norm of the error matrix and the normalized SINR through numerical simulations. The simulation results show that the new method outperforms existing methods in both accuracy and robustness.