Sensor Fusion of Ultra-Wideband and Bluetooth 5.1 for Precise Indoor Positioning

Abstract

This paper presents a hybrid indoor positioning system that combines the high-precision distance measurement (Time of Flight) of ultra-wideband (UWB) with the directional angle estimation (Angle of Arrival) of Bluetooth 5.1 (BLE) via sensor fusion. As individual radio-based systems often reach their limits in complex environments due to multipath propagation and non-line-of-sight (NLoS) conditions, the proposed approach utilizes a modular hardware architecture and multistage signal preprocessing pipelines on a central server. For the BLE subsystem, the MUSIC algorithm is extended to include calibrations for DC offsets and phase drifts, while UWB NLoS errors are corrected using real-time capable 3D ray-casting and empirical attenuation models. The evaluation demonstrates the high robustness of the sensor fusion: In severely obstructed NLoS scenarios, the root-mean-square error (RMSE) in positioning could be reduced by 82 % (from 1.25 m to 0.22 m) through the correction procedures. The system effectively compensates for the weaknesses of the individual technologies and enables highly precise 3D localization.