Weight estimation in STAP for better SINR using AIC

Abstract

Space Time Adaptive Processing (STAP) has been proposed to suppress the clutter in airborne radars. In real situation, the sample covariance matrix is used to estimate the optimum weight to suppress the clutter, resulting its improved Signal to Interference plus Noise Ratio (SINR). The weight is estimated by an eigenvector set of the covariance matrix of received signals. Conventionally, all of the eigenvectors are included in the weight, assuming all eigenvectors belong to the signal subspace, which means the eigenvectors derive from the clutter. However, when the weight includes some eigenvectors belonging to the noise subspace, the resultant SINR is degraded, because the components other than the clutter are subtracted from the received signals. Therefore, in order to avoid this degradation in STAP, we have proposed the new weight estimation method which selects eigenvectors included in the weight based on Akaike Information Criteria (AIC). The evaluation for our proposed method was carried out using the data measured at the Sea of Japan by an airborne side-looking radar, which include a ship as a target and sea clutters in real environment. The results show that the proposed method which limits eigenvectors included in the weight achieves better SINRs than the conventional method.