Non-line-of-sight Target Detection With MIMO Millimeter Wave Radar Based On Multi-path Utilization

When the target is blocked by buildings and other obstacles,the traditional radar line-of-sight detection method is ineffective,but it can use electromagnetic wave reflection path propagation to realize non-line-of-sight target detection.In recent years,non-line-of-sight target detection has been widely applied in the fields of intelligent home,intelligent driving,urban reconnaissance and perception,and has become a research hotspot.Aiming at the detection problem of non-line-of-sight targets in masking scenes,this thesis studies the multi-path propagation model,non-line-of-sight target location,tracking method and vital feature estimation in typical non-line-of-sight scenes based on multi-input and multi-output millimeter-wave radar.The main research contents are as follows:1.Aiming at the typical L corner scene,the reflected multipath propagation characteristics of electromagnetic wave in this scene are analyzed,and the reflected multipath propagation model and echo signal model are established.At the same time,the principle of target detection based on millimeter wave radar is studied,including target range,speed and angle estimation principle,which lays a theoretical foundation for the subsequent target location and tracking.2.Aiming at the problems of stationary target location and respiration parameter estimation in L non-line-of-sight scene,combined with the reflection multipath propagation model in this scene,the echo signal preprocessing,constant false alarm rate(CFAR)detection method and the minimum variance distortionless response(MVDR)angle estimation method based on millimeter wave radar are studied,and implement target location.On this basis,the estimation method of breathing parameters of stationary target based on Fourier transform is studied,and the target breathing number is estimated.3.Aiming at the problem of moving target tracking in the L non-line-of-sight scene,by utilizing the reflected multi-path signal of millimeter-wave radar,a non-line-of-sight target tracking method based on Trace Before Detect based on Dynamic Programing(DPTBD)is proposed.Through multi-frame data joint processing,the detection range of nonline-of-sight targets is increased while the detection probability is guaranteed,and the robust tracking of non-line-of-sight targets is realized.All the above algorithms are verified by the experimental results,and it is proved that the proposed algorithm has a good effect on non-line-of-sight target detection.