SAR/GMTI Vehicle Target Classification Based On Micro-doppler Feature

Micro-motion produces modulation of radar echo of a moving target,and this phenomenon is called micro-Doppler effect.The micro-Doppler feature,containing the structure and movement patterns of moving target,is unique and stable,therefore becomes valuable cue to identify the target.Because there are significant different on the function and threaten degree between the track-laying vehicle and the wheeled vehicles on the battlefield,classification of two type of vehicle is valuable.By analyzing the frequency of the modulated echoes,it has been found that significant differences between different targets exist,thus proving an indication that a micro-Doppler signature can be used as the basis for target classification or recognition.AS for SAR/GMTI radar system,two kinds of EEMD-based vehicle target recognition methods are proposed: one is based on the energy distribution feature and the other is based on the extended micro Doppler features.The main works of the paper are as follows:1、In this study,we overcome the mode mixing problem by applying Ensemble Empirical Mode Decomposition(EEMD)to extract the micro-Doppler signal.The simulation results show that by adding Gaussian white noise,EEMD decomposition overcomes the model mixing problem.2、A new vehicle target recognition method based on energy distribution characteristics with EEMD is proposed.The EEMD algorithm is employed to obtain the IMF components,the effective energy between the IMF components than is calculated,and three typical energy distributions are extracted as the feature vectors.The simulation results show that the vehicle target recognition method based on energy distribution characteristics with EEMD will achieve a high recognition rate for large velocity range and azimuth range.3、A new vehicle target recognition method based on extended micro Doppler features with EEMD is proposed.In order to select some specific intrinsic mode functions(IMF),which are most correlative to the micro-Doppler signal,the correlation analysis is then carried out.The selected IMFs are used to extract four features based to their differences between the wheeled and the tracked vehicle signals.Four features are extracted in this paper: the energy in high frequency band of IMF1,the volatility in high frequency band of IMF1,the spectrum maximal magnitude of IMF2,the dispersion among IMFs based on energy entropy.The simulation results show that vehicle target recognition method based on extended micro Doppler features with EEMD is superior than vehicle target recognition method based on energy distribution characteristics with EEMD on recognition effect.