摘要

For airborne conformal array radar, it is difficult to suppress clutter using conventional space-time adaptive processing (STAP) methods because the geometry results in range-dependent clutter and nonuniform spatial steering vector. In order to improve the performance of conformal array clutter suppression, a knowledge aided method based on semiparametricisparse iterative covariance-based estimation STAP, named with KASPICE-STAP, is proposed here. The KASPICE-STAP method requires the knowledge of the clutter ridge spread of the testing range cell. Based on this knowledge, we can construct a dictionary matrix which consists of the clutter space-time steering vectors. Subsequently, by using the dictionary matrix, a relatively accurate covariance matrix of clutter plus noise can be calculated with the KASPICE-STAP method. The proposed approach can work efficiently in complex environments only with the data of cell under test. Compared with other existing sparse recovery STAP methods, the proposed method has global convergence properties and it does not require making any difficult selection of hyperparameters. The improvement factor (IF) curves and range-Doppler images demonstrate the effectiveness of the proposed method.