Research On Vehicle-Mounted Millimeter-Wave Radar Signal Processing Algorithm

With the advancement of science and technology,intelligent vehicles have become an inevitable trend for the development of automobiles in the future.As one of the key sensors of intelligent vehicles,vehicle-mounted radar is getting more and more attention.In common vehicle-mounted radars,millimeter-wave radar has unique advantages over laser radar,and it is indispensable in car collision avoidance systems.In order to improve the detection performance of the vehicle-mounted millimeter-wave radar,in addition to constantly enhancing the hardware performance of the radar system,the research of signal processing algorithms has been receiving much attention.The most commonly adopted system for vehicle-mounted millimeter-wave radar is the linear frequency modulated continuous wave(FMCW)system.For the millimeter-wave radar of the FMCW system,the signal processing structure of “Beating-Frequency Spectrum analysing-MTD” is usually adopted,that is,by performing spectrum analysing of the beat signal to achieve target detection and target parameters(e.i.,range,speed and angle)estimation.And by adopting moving target detection(MTD)technology it can overcome the range-velocity coupling problem of FMCW radars and achieve multi-target information pairing.In addition,the introduction of beamforming technology can effectively enhance the target signal and suppress interference,clutter and noise,thereby improving the target detection performance in complex scenarios with multiple targets,multiple interferences,strong clutter and strong noise and alleviating the problem of the limited reach of the FMCW radar.In this thesis,based on the signal processing structure of “Beating-Frequency Spectrum analysing-MTD” for the FMCW-based vehicle-mounted millimeter-wave radar,with introducing the beamforming technology,the signal processing structure of “Beating-Beamforming-Fourier Spectrum analysing-MTD” is obtained and under it the signal processing algorithms are studied.Firstly,the overall scheme and principle of FMCW system-based millimeter-wave radar signal processing algorithm are studied.The main contents include: on the one hand,the basic principles of target's range,speed and angle estimation and other basic principles are discussed through the analysis of multi-periodic symmetrical triangular wave signals.And the principle of frequency domain constant false alarm detection(CFAR)algorithm to implement target detection and the principle of MTD-velocity pairing to solve range-velocity decoupling and to achieve multi-target pairing are also studied;on the other hand,under the array signal model,adaptive beamforming algorithm is studied,which includes receive beamforming and transmit beamforming and joint transmit-receive beamforming to enhance the target and suppress interference,clutter and noise.Secondly,through MATLAB simulation,the effectiveness of the algorithms of the signal processing algorithm scheme are verified,where the main algorithms are frequency domain CFAR,target parameters(e.i.,range,speed and angle)estimation,MTD-velocity pairing,and transmit/receive beamforming.Finally,based on the DSP processor,the implementation of the FMCW vehicle-mounted millimeter-wave radar signal processing algorithm is studied,and some major algorithm modules are implemented,including CFAR detection algorithm and range and velocity estimation and pairing.