Indoor Detection and Tracking of People Using mmWave Sensor

This paper proposes a new indoor people detection and tracking system using millimeter-wave (mmWave) radar sensor. Firstly, systematic approach for is presented—a static clutter removal algorithm used removing mmWave data’s points. Two efficient clustering algorithms are to cluster identify in scene. The recursive Kalman filter with data association track multiple simultaneously. Secondly, fast designed based on our proposed algorithms. method lightweight enough scalability portability, we can execute it real time Raspberry Pi 4. Finally, the validated by comparing Texas Instruments (TI) system. system’s experimental accuracy ranged from 98% (calculated misclassification errors) one person 65% five people. average position errors at positions 1, 2, 3 0.2992 meters, 0.3271 0.3171 respectively. In comparison, had an ranging 96% 45% 0.3283 0.3116 0.3343 method’s advantage demonstrated terms of accuracy, computation time, scalability.