Robust parametric detection in non homogeneous clutter

In this paper we address the problem of robust adaptive parametric radar detection in non-homogeneous clutter. Interfering targets and clutter edges in secondary cells are regarded as outliers. Classical estimators of parametric models are not sufficiently robust in such situations. We consider stationary segments, where the clutter is modelled as an autoregressive process, we apply a robust filter and we estimate the autoregressive model to construct the parametric detector.