Classification of Human Motion Status Using UWB Radar Based on Decision Tree Algorithm
Ultra-wideband radar signals have broad application prospects in ultra-close-range detection due to its high resolution, strong penetrability and strong anti-interference. Ultra-wideband radar is an ideal detection method in non-contact medical monitoring scenarios. Based on UWB radar, the time domain sliding window feature extraction method and decision tree classification algorithm are used to classify and recognize the human motion state behind the obstacle. The decision tree is generated by inputting the sliding window feature extraction result of the radar echo data into the decision tree algorithm. The experimental results show that the proposed feature extraction algorithm and decision tree algorithm can classify the motion state of the human target behind the obstacle, such as the human walking in the radial direction, walking in the tangential direction, and standing in the ground.