Indoor Positioning System via Bluetooth Low Energy and mmWave Radar Sensor Fusion

Indoor positioning is vital for many applications such as asset tracking, context-aware automation, and human-machine interaction, yet achieving sub-meter accuracy remains challenging. This paper proposes a sensor fusion approach that integrates Bluetooth Low Energy (BLE) and millimeter-wave (mmWave) radar to improve positioning performance. We show that, due to the high variance of the BLE measurements, sensor fusion alone does not considerably improve position accuracy over BLE and mmWave. To mitigate that issue, a sliding window median filter is applied to BLE measurements to enhance stability, and the refined data are fused with mmWave estimates using a track-to-track method. Experimental results demonstrate that the proposed approach reduces positioning error by 42.83% and 25.40% compared to BLE and mmWave alone, respectively.