Yuan Yang The Ohio State University
Charles K Toth The Ohio State University
Zoltan Koppanyi The Ohio State University

15F

Abstract:

With the growing number of multi-sensor equipped smart mobile devices, fingerprinting-based localization techniques, including radio frequency (RF) based signal fingerprint, inertial sensor based motion fingerprinting and content-based image retrieval (CBIR) based visual fingerprinting, are nowadays popular for many indoor location-based services (ILBS), due to they are more robust and have lower computational cost as compared to other signal-based and sensor-based indoor localization techniques such like Time of Arrival (ToA), Time Difference of Arrival (TDoA) and Simultaneous Localization And Mapping (SLAM). Fingerprinting techniques estimate devices’ location by matching the on-line observation of fingerprint with the recorded fingerprint from predefined fingerprint map, or database. Under this concept, all standalone fingerprinting techniques have fingerprint variance and efficient map updating problems due to the environment changing. To deal with these problems, we propose a new hybrid fingerprinting system which organically fuse Received Signal Strength (RSS) based Wi-Fi fingerprinting and CBIR based visual fingerprinting to adapt indoor environment changing. The system was tested in a real-world experiment conducted with indoor office environment.

15:45 An Adaptable Hybrid Fingerprinting System for Indoor Localization and Fingerprint Map Updating, Yuan Yang