Seamless and Efficient Indoor–Outdoor Handover Using GNSS and Bluetooth Low Energy : Design, Implementation, and Evaluation of a Hybrid Positioning System

Sammanfattning

Hybrid positioning systems (HPS) are becoming increasingly important, as modern applications require reliable positioning in both indoor and outdoor environments. Global Navigation Satellite Systems (GNSS) provide accurate outdoor positioning but perform poorly indoors, while Bluetooth Low Energy (BLE) technologies are more suitable for localization in indoor environments. Therefore, a major challenge is to achieve a seamless and stable handover between these systems during transitions between indoor and outdoor areas. The purpose of this study was to implement and evaluate a handover mechanism for an HPS based on GNSS Real-Time Kinematics (RTK) and BLE Angle of Arrival (AoA). The study focused on handover reliability, adaptive BLE scanning for improved energy efficiency, and the use of predictive BLE signal features during transitions. An experimental and quantitative approach was employed, involving transition experiments conducted in controlled indoor-outdoor areas. A confidence-score-based handover mechanism, together with hysteresis thresholds, was implemented to determine switching points between the GNSS and BLE subsystems. The results showed that the proposed handover mechanism maintained stable transitions with a maximum measured handover decision latency of 105 ms in all evaluated sessions. The adaptive BLE scanning approach reduced scanning activity without significantly affecting handover stability. Additionally, the results indicated that combining several BLE signal features, specifically Received Signal Strength Indicator (RSSI), RSSI variance, and RSSI slope, improved the reliability of detecting degrading signal conditions during transitions. The study demonstrates that an HPS based on commercially available hardware can achieve stable indoor-outdoor handover performance under controlled conditions. Furthermore, the results indicate that predictive BLE behavior and confidence-score-based decision mechanisms may improve the robustness of future HPS implementations.