A CFAR Algorithm Based on Monte Carlo Method for Millimeter-Wave Radar Road Traffic Target Detection

The development of Intelligent Transportation Systems (ITS) puts forward higher requirements for millimeter-wave radar surveillance in the traffic environment, such as lower time delay, higher sensitivity, and better multi-target detection capability. The Constant False Alarm Rate (CFAR) detector plays a vital role in the adaptive target detection of the radar. Still, traditional CFAR detection algorithms use a sliding window to find the target limit radar detection speed and efficiency. In such cases, we propose and discuss a CFAR detection method, which transforms the Monte Carlo simulation principle into randomly sampling instantaneous Range–Doppler Matrix (RDM) data, to improve the detection ability of radar for moving targets such as pedestrians and vehicles in the traffic environment. Compared with conventional methods, simulation and real experiments show that the method breaks through the reference window limitation and has higher detection sensitivity, higher detection accuracy, and lower detection delay. We hope to promote the detection application of millimeter-wave radar in road traffic scenes