Indoor 3-D Localization Based on Heterogeneous Sensing Data Fusion

To address the challenges of better balancing the cost and positioning accuracy in indoor localization and mitigating the performance degradation due to signal blocking and multipath propagation in complex three-dimensional (3D) space, this paper presents a 3D localization system that integrates heterogeneous sensing data. Specifically, by combining the cost-effective Bluetooth Low Energy (BLE) and high-ranging-precision Ultra-Wideband (UWB) technologies, the developed scheme realizes 3D indoor localization using no more than two UWB base stations (BSs). The system can dynamically switch between the single BS and dual BS modes according to the changes in the target mobility and environmental interference, enhancing its adaptability to various scenarios. The positioning results are found through solving a geometrically constrained optimization problem. A BS selection mechanism is further introduced to improve the performance. Real-world experimental results demonstrate that the proposed system has significantly enhanced indoor localization accuracy, with improvements ranging from 23.3% to 42.1% compared to conventional methods. Its deployment requires only at most two UWB BS and a limited number of BLE hotspots, which demonstrates its significant potentials in practical applications.