Research On Micro-Doppler Feature Modeling And Analysis Technology Of Radar Targets

With the rapid development of military technology,all kinds of motion targets and small UAV with threat have posed a serious challenge to the ability of traditional radar identification.Radar identification based on micro Doppler feature extraction has the problem of low frequency resolution which can't identify the target accurately.Thus,how to improve the frequency resolution of micro Doppler signal has become a research hotspot in the target identification area.This thesis mainly studies on high resolution of micro Doppler feature extraction methods.Main works are as follows:(1)Studying the micro movement of all kinds of targets and establishing the motion model.The micro Doppler echo signal is obtained through the motion model which is a good foundation for the later analysis of the micro Doppler feature extraction methods.(2)The conventional micro Doppler feature extraction methods provide low frequency resolution which is caused by fixed window length and time-frequency focusing problems.S transform is used to address this problem,and the simulation result showed this method did improve the frequency resolution of time-frequency analysis.(3)S transform has two problems,one is the loss of signal energy caused by signal truncated,the other is low time-frequency concentration.In order to solve these problems,this paper proposed a hybrid time-frequency analysis method which combines the idea of S transform and linear canonical transform.In the proposed method,S transform is mapped into the linear canonical domain,and by setting the relevant parameters of linear canonical transform can make the equivalent window match the frequency of signal very well.Simulation result showed this method improved the frequency resolution of time-frequency analysis a lot.(4)Signal is always submerged in the noise.In order to address this problem,this paper proposed a method which is multiple autocorrelation.When the signal is being separating from the noise,strengthening the signal at the same time.Simulation results show that this method is effective in extracting useful noise from micro Doppler signal,and the signal is enhanced,the time-frequency analysis results show that the micro Doppler feature is obvious,so the low SNR extraction of micro Doppler features is being realized.