Application of Compressive Sensing to anlge-Doppler processing without Knowledge-Aid

摘要

The Compressive Sensing (CS) is a mean to solve simultaneously detection and estimation problems. It has been applied mainly in source Direction of Arrival (DOA) estimation problems where the amount of data to be handled is reasonably small. This is not the case in Knowledge-Aid Space Time Adaptive Processing (KA-STAP) which is widely used for ground clutter suppression and targets detection in airborne radar for the extremely large amount of data. We develop an efficient algorithm which can provide two supervisors: one is that target can be separated from ground clutter in the condition of small amount of training data; the other one is that the number of sources is not necessary.