With the widespread deployment of WiFi infrastructure, WiFi RSSI fingerprinting has been extensively explored for indoor localization. However, conventional fingerprinting approaches require labor-intensive data collection and are often constrained by limited sampling rates in practical deployments. To address these limitations, we propose WiMU, a real-time indoor localization system that integrates WiFi and inertial measurement unit (IMU) data to enhance the real-time performance and accuracy of localization. WiMU operates on commodity WiFi infrastructure without the need for additional hardware, leveraging crowdsourced user trajectories to learn spatial representations of access points (APs). These representations can be fine-tuned with minimal labeled data to support effective localization. Extensive evaluations show that WiMU outperforms three state-of-the-art methods, significantly reducing the overhead while maintaining high positioning accuracy. Notably, it achieves an average localization error of 5.247 meters using no more than 100 fingerprints with known locations. WiMU has also been successfully deployed in real-world environments, including a university campus and a warehouse, demonstrating its practical effectiveness for real-time indoor localization.
Yang et al. (Tue,) studied this question.