Clutter Suppression Methods For Airborne Radar Moving Target Indication

Moving target indication play an important role in Early Warning and battlefield surveillance, which includes air moving target indication and ground moving target indication. Doppler offset resulting from high velocity of air moving target makes it deviate from main clutter in frequency, and then air moving target can be detected in sidelobe clutter. When coherent time is short, poor Doppler resolution gives rise to results that main clutter is broaden and weak target is covered. A projection matrix method is presented for this problem which detect moving target through eliminating main clutter using project matrix. As for ground moving target, it is often covered by main clutter due to its low velocity. In this case we should adopt space and time adaptive process to suppress clutter, however, many practical factor can induce into clutter data non-independently and non-identically distributed (non-iid), thus, performance degradation happens when adaptive weight is computed by covariance matrix estimated from nonhomogeneous secondary data. Therefore all sorts of methods for compensating nonhomogeneous effect should be adopted. In this dissertation, the effect of some common phenomena on detection performance is discussed and some common methods for nonhomogeneous clutter suppression are presented including local training method, general inner product method and direct data domain method. Finally simulation results validate the effectiveness of the methods presented in this dissertation.