Reduced Dimension STAP Based on Sparse Recovery in Heterogeneous Clutter Environments
Wei Zhang,Ruixue An,Ningyu He,Zishu He,Huiyong Li
DOI: https://doi.org/10.1109/taes.2019.2921141
IF: 3.491
2019-01-01
IEEE Transactions on Aerospace and Electronic Systems
Abstract:For airborne-phased array radar systems, space-time adaptive processing (STAP) is supposed to be a crucial technique for improving target detection performance in the strong clutter background. However, practical application environments are always heterogeneous and have offered a severe challenge to the implementation of STAP. To address this reality, a data-dependent reduced dimension STAP approach based on sparse recovery (SR) is proposed in this paper. To take a full account of the heterogeneous environments, we consider the extremely heterogeneous case in which only one single snapshot is available. There is no doubt that, compared with the clutter covariance matrix (CCM) calculated directly by a single snapshot, the SR technique can provide a more accurate estimate. However, we should come to realize that this estimation is not accurate enough to adaptive processing according to the presented simulation results in lots of literature, and it has been demonstrated that the performance of traditional SR-based STAP degrades dramatically when only a snapshot is available. In the proposed approach, the CCM estimated by SR technique is utilized to design the reduced dimension transformation matrix rather than to calculate adaptive weights as the traditional SR-based STAP. The relatively accurate CCM can provide better support for the design of reduced dimension transformation matrix. From the simulation results, the proposed approach can achieve great performance of clutter suppression and target detection with only a single snapshot compared with several typical STAP algorithms. It is worthwhile pointing out that the proposed approach can also be applied when multiple snapshots are available and the performance improves with increasing number of available snapshots.