A High Performance Framework for Radar Clutter Suppression, Target Tracking and Identification

We propose a high performance system for clutter suppression, target tracking, and identification. First, we propose a novel clutter rejection approach that applies an adaptive subspace projection algorithm. As compared to conventional clutter suppression methods, our adaptive clutter rejection algorithm is independent of target detection methods and does not assume any target signal model. Our method is applicable to any kinds of clutter, including infrared, cloud, etc. Second, we propose to apply adaptive chirplet transform to extract the time-varying Doppler frequencies in the target signals. Conventional target detection methods are based on Fourier transform method and do not work for time-varying Doppler frequencies. Our simulations showed that we can still detect dim targets even when the raw data have a signal-to-clutter-ratio (SCR) of less than -50 dB, which is a significant result. Third, we propose to apply latest ballistic target tracking algorithm to track multiple moving targets. Our algorithm is computationally efficient and yields near optimal tracking performance. Fourth, based on the tracking results and target shape information, it is possible to differentiate real missiles from decoys. In Phase 1, we plan to implement the proposed framework and perform extensive simulations to validate its performance. In Phase 2, we will develop a real-time system.