Research On Target Detection Algorithm Based On 77GHz Millimeter Wave Radar

Millimeter wave radar has a wide development prospect in the field of target detection because of its unique advantages such as high resolution,high precision,little environmental influence and easy to protect privacy.This paper studies the target detection algorithm of 77 GHz millimeter-wave radar in three aspects of indoor personnel detection and tracking,dynamic gesture recognition and respiratory and heartbeat sign detection.The main work is as follows:The indoor personnel detection and tracking algorithm based on millimeter wave radar is studied.In order to overcome the interference from the surrounding static objects,MTI static interference elimination algorithm is proposed to suppress the static clutter in the echo.In order to further filter clutter and noise interference and filter out the target of our concern from multiple wave peaks,TM-2D-CFAR algorithm is designed to carry out two-dimensional constant false alarm target detection(CFAR)to realize the stable detection of multiple targets.Aiming at the indoor personnel tracking problem,this paper proposes a target tracking algorithm based on extended Kalman filter to realize the nonlinear tracking of detection personnel.Dynamic gesture recognition algorithm based on millimeter wave radar is studied.In order to extract gesture features,RDM was obtained by two-dimensional FFT processing of IF signals,and various statistical features were generated based on RDM,which were taken as original features.In order to solve the problem that the original feature has too high dimension,Relief F algorithm was proposed to filter the features,and the redundant features were further eliminated by clustering algorithm to obtain the feature subset highly relevant to gesture recognition.In addition,the DAG-SVM multi-classifier is designed to classify gestures based on the features selected after the traditional SVM algorithm,and the recognition accuracy is high.The respiratory and heartbeat sign monitoring algorithm based on millimeter wave radar is studied.Taking respiration rate and heart rate as vital signs detection indexes,the vital signals and clutter were modeled and analyzed,and an adaptive filter was designed to effectively filter clutter components from the echo signals.In view of the difficulty in detecting thoracic displacement,an algorithm for detecting thoracic displacement by phase change is proposed.In order to solve the problem that the frequency spectrum of respiratory signal and heartbeat signal is alias and difficult to separate,an LMS adaptive filter structure is designed to filter the respiratory harmonic component from the heartbeat signal,and two elliptical filters are designed to separate the respiratory heartbeat signal.This paper improves the traditional time domain peak search and combines the frequency domain peak search to process the respiratory heartbeat signal,so that the estimation result of respiratory heartbeat rate is more accurate.MATLAB is used to simulate the above algorithm,and IWR1642 platform is used to verify the feasibility of the algorithm.