The Practical Real-time Radar Processing for Obstacle Detection on The Railway Track
Tossaporn Srisooksai, Satoshi Nishida · 2024
This paper considers the MIMO radar processing to detect and track the obstacles on the railway track, where a high rate of real-time processing is necessary. Since there exist many obstacles located outside the detection area in this environment, the typical constant false-alarm rate detector usually detects all of such undesired obstacles. This requires a higher computation workload in the next steps of processing (Angle Fast Fourier transform, clustering, and tracking) and hinders the high rate of real-time radar processing. Therefore, this work proposes a detection approach that can suppress the power of rangevelocity spectrum bins caused by the reflection waves generated by the obstacles existing outside the detection area to reduce the consequent computation workload after this detection block. The measurement result verified such merit.