Efficient Discrimination of Ballistic Targets With Micromotions

The micro-Doppler phenomenon in the echo signal received from a ballistic target (BT) with micro-motion is commonly used to discriminate BTs such as warheads and decoys. The joint time-frequency (JTF) analysis of the echo signal has been considered as useful two-dimensional (2-D) information in BT discrimination, which generally requires a framework based on the processing of the 2-D JTF image with various conventional feature extraction techniques. However, these techniques are inefficient for time-critical BT discrimination task due to the complicated 2-D image processing. In this paper, we propose new echo signal models to formulate the fundamental difference between the micro-motions of warheads and decoys, leading to a novel BT discrimination framework via new feature extraction paradigm and multi-aspect fusion concept. The most attractive attribute of this framework is that it can provide substantial savings with regard to computational resources as well as robustness to noise. The experimental results illustrate that the proposed discrimination scheme shows considerable promise for application in real-time BT discrimination.