Research On The Imaging Method Of Close Range Targets In Vehicle Mounted Millimeter Mave Radar

In recent years,the incidence of traffic accidents has sharply increased with the large number of cars entering ordinary people’s homes.The development of advanced driving assistance systems(ADAS)can greatly reduce the incidence of accidents and reduce the severity of collisions.Vehicle mounted millimeter wave radar has become an indispensable sensor for ADAS due to its low hardware cost,long detection distance,strong weather adaptability,and ability to penetrate smoke,dust,rain,snow,and fog.The use of onboard millimeter wave radar to detect three-dimensional(3D)point cloud images of distance,velocity,and angle of targets in front can provide a data foundation for subsequent clustering,tracking,and sensor data fusion.This article systematically studies the imaging methods for close range targets of on-board millimeter wave radar under the 77 GHz frequency modulated continuous wave(FMCW)time division multiplexing(TDM)multi input multi output(MIMO)millimeter wave radar system,mainly including Doppler phase offset compensation algorithm,velocity expansion algorithm,high-precision angle measurement method,and imaging verification system design.The specific details are as follows:(1)In response to the problems of Doppler phase deviation error and velocity ambiguity in the TDM MIMO radar system,the reasons for Doppler phase deviation error and velocity ambiguity were first analyzed.Then,based on this,a Doppler phase deviation compensation algorithm based on velocity estimation,a Chinese remainder theorem(CRT)based algorithm,and a velocity expansion algorithm based on peak detection were studied.The simulation and experimental results show that after Doppler phase offset compensation for multiple virtual receiving array elements,the angle estimation error is corrected.The speed expansion algorithm based on CRT supports up to 3 times the speed expansion,and the speed expansion algorithm based on peak detection supports up to 2 times the speed expansion.(2)After Doppler phase offset compensation and velocity expansion,two-dimensional(2D)point cloud images of distance and velocity were obtained,and additional angle information was needed to obtain 3D point clouds.However,existing methods have the problem of low angle accuracy due to off grid errors.Therefore,by utilizing the statistical characteristics of the Maximum Likelihood(ML)criterion,a high-precision angle measurement method combining Iterative Sparse Learning via Iterative Minimization(SLIM)and ML estimation is proposed for point cloud imaging.This method first utilizes the SLIM method for power estimation to obtain the initial target angle value;Then,the ML cost function is minimized to refine the Direction of Arrival(DOA),thereby breaking through the limitations of grid partitioning on angle estimation accuracy.The simulation and experimental results show that the accuracy of the angle dimension of the point cloud image obtained based on this method can reach 0.1 degrees while ensuring high resolution.(3)Based on the previous research,a vehicle mounted millimeter wave TDM MIMO radar imaging verification system was designed.The system uses Texas Instruments(TI)millimeter wave radar sensor AWR1843 and data acquisition board DCA1000 as the hardware platform,and TI’s mm Wavestudio software and an upper computer interface written based on MATLAB as the software platform.The system controls mm Wavestudio to configure radar parameters and start radar work through the upper computer.DCA1000 receives IF data collected by AWR1843 and uploads it to the upper computer.The upper computer analyzes the data,processes the signal and displays the interface according to the received binary bin file and the configuration information Logfile file.The experimental results show that the system has the functions of Doppler phase deviation compensation,speed expansion,and improving angle accuracy.The upper computer has the advantages of convenience,intuition,and ease of use.