Robust Classification Scheme for Airplane Targets With Low Resolution Radar Based on EMD-CLEAN Feature Extraction Method

Abstract

A novel classification scheme is proposed to categorize airplane targets into three kinds, i.e., turbojet aircraft, prop aircraft, and helicopter based on the jet engine modulation (JEM) characteristics of their low resolution radar echoes. From the pattern classification viewpoint, the low-dimensional feature vector is extracted via a two-step feature extraction algorithm based on empirical mode decomposition method and CLEAN technique. The feature extraction method can separate the fuselage component and JEM component in the target echo, and sufficiently utilize the information within or between the two components to extract the discriminative features for the three kinds of aircrafts. In addition, because the noise level of a test sample is usually different from those of the training samples in the real application, a simple and efficient preprocessing method is proposed for the classification stage to denoise the received test sample. Experimental results based on the simulated and measured data are presented, including the performance analysis for different dwell time, pulse repeat frequency (PRF) or signal-noise-ratio (SNR) cases and comparison with some related methods.