Research On LFMCW Radar Slow And Weak Target Detection Technology

With the advantages of small volume,light weight,high range resolution,no blind range and low probability of interception,liner frequency modulated continuous wave(LFMCW)radar has been paid more and more attention recently.The key technologies and application develop rapidly.The detection of slow and weak targets have gradually developed into a popular research aera.Track before detect(TBD)algorithm is proposed for this task.In this thesis,focused on the LFMCW radars,target detection technique for slow and weak targets is studied,which includes the technique of long time coherent integration,the TBD algorithm based on dynamic programming and constant false alarm rate(CFAR)detection in multiple targets situations.The research works are as follows:1.Focused on the slow and weak radar targets,the novel motion compensating waveforms algorithm(MCW)and Radon-Fourier transform(RFT)are introduced respectively for radar long time coherent integration with across range units(ARU)range walk.The MCW frequency and bandwidth agility waveforms are proposed for the targets with ARU range walk.The matched filter output of RFT algorithm for LFMCW radar is analyzed,which validates the correctness of RFT algorithm for LFMCW radar.2.Focused on the slow and weak targets with non-uniform velocity,the TBD algorithm based on dynamic programming for LFMCW radar is studied.The improved weighted DP algorithm are introduced,which can achieve better performance.This thesis proposes the application of DP algorithm in radar weak targets,and summarizes the target representation,system model and algorithm steps.The determination of value function threshold is analyzed and the correctness of the algorithm is verified by the measured data of LFMCW radar.3.Considering the characteristic that the statistic skewness in Weibull distribution is a constant,an improved CFAR processor called SK-CFAR based on skewness for multiple targets situation in LFMCW radar Weibull clutter environment is proposed to adaptively remove the interfering targets from the reference windows and improve the detection performance.The Monte-Carlo simulation proves that the SK-CFAR processor actually performs better than the traditional ones and the detection of interfering targets numbers and algorithm complexity has been greatly improved.