STRUCTURED COVARIANCE MATRIX ESTIMATION FOR THERANGE-DEPENDENT PROBLEM IN STAP

摘要

We propose a method to compute an estimate of the clutter-plus-noise covariance matrix in bistatic radar configura-tions. The estimation is based on the computation of theclutter scattering coefficients based on a single data snap-shot at each range using a model of the received signal.The covariance matrix of the data is modeled as a struc-tured covariance matrix with the scattering coefficients asunknown parameters. The method is based on the compu-tation of the maximum likelihood. We use the Expectation-Maximization method as estimation benchmark. Since theproblem is ill-posed, regularization is mandatory. This reg-ularization is performed by spatial smoothing. The methodwe propose, unlike the Expectation Maximization, is notiterative and is thus less computationally demanding. The obtained covariance matrix estimate is used tocompute the matched filter in order to perform target detec-tion. The performance of the proposed estimation methodis evaluated in terms of signal to interference-plus-noise ra-tio (SINR) losses and is found to be almost indistinguish-able from the performance of the clairvoyant case.