Integrated WiFi/MEMS Indoor Navigation Based on Searching Space Limiting and Self-calibration

Indoor navigation has been increasingly popular over the last few years. However, it still faces plenty of challenges and remains a conundrum. This paper proposes a novel improved WiFi/MEMS integration solution for indoor navigation. In WiFi fingerprinting scheme, a novel searching space limiting method is originally presented and associated with a mean filter to improve computation efficiency and positioning accuracy, compared with the traditional weighted K-nearest neighbors method. In pedestrian dead-reckoning part, an attitude determination extended Kalman filter with correlated process and measurement noise is presented to obtain an accurate long-term heading and the average positioning error decreases significantly as a result. Furthermore, the self-calibration Kalman filter approach is introduced into indoor navigation field in WiFi/MEMS integration stage and a novel Kalman filter system is originally designed to fuse the information effectively. The navigation performance of the proposed WiFi/MEMS algorithm has been validated by indoor experiments, and the average positioning error is less than 0.6 m when the number of selected APs is optimal.