Radar clutter suppression based on sαs fractional autoregressive model

In correlated non-Gaussian clutter backgrounds, serious degradation occurs in traditional clutter suppression method. A new clutter suppression method was proposed based on Symmetric alpha stable (SαS) fractional autoregressive model. The SaS fractional autoregressive model was considered as a stochastic processed in which a fractional autoregressive system was driven by a white SaS noise, and the clutter suppression Alter was established by using the model parameters which was estimated based on the Fractional lower order covariance (FLOC). The SaS fractional autoregressive model can effectively describe the non-Gaussian characteristics as well as the long and short correlation characteristics of the clutter, and by using FLOC, more accurate parameters can be estimated. Simulations and real data results show that the proposed method obviously outperforms the traditional clutter suppression method under correlated non-Gaussian clutter backgrounds.