Multipath Model and Ghost Suppression for Millimeter Wave Traffic Radar

Millimeter-wave traffic radar, known for its advantages of all-day, all-weather, and long-distance monitoring, plays a critical role in traffic safety. However, the presence of multipath ghosts has persistently degraded the performance of traffic radar systems. In this paper, we establish a radar echo signal model by analyzing the propagation paths of multipath signals. Key features, including the distance between multipath ghosts and road guardrails, as well as the continuity of detection point clouds, were systematically analyzed. Based on these identified features, we propose a neural network-based method for ghosting suppression. Empirical data analysis validates the accuracy of target detection and tracking by millimeter-wave traffic radar. The results demonstrate that the utilization of the proposed eight features effectively mitigates the presence of multipath ghosts, enabling precise positioning and accurate tracking of vehicles.