Abstract: Currently, missileborne radar usually has ΣΔbeam. Because of the electromagnetic coupling field between superlowaltitude targets and the surface of the ground or the sea, the false image target will be included in the training samples. Thus, the target cancelation will happen and the performance of moving target detection and clutter covariance matrix estimation will degrade dramatically. To deal with this problem, a robust spacetime adaptive processing (STAP) method based on the joint magnitude and phase constraints imposed on and around the target is proposed. As a result, the mainlobe of sum beampattern is well maintained, the performance loss due to intertarget contamination can be alleviated, and the Doppler estimation performance is also improved in terms of accuracy and robustness. Moreover, the clutter has range dependence for forward looking missileborne radars in short range gates, and the method based on least square can be used to deal with this problem. Simulation examples are presented to demonstrate the effectiveness of the proposed method.