A 2-stage hybrid position estimation framework in RF fingerprint WPS

Abstract

The rapid growth of mobile communication and the proliferation of smart phones have drawn significant attention to Location Based Services (LBS). One of the most important factors in the vitalization of LBS is the accurate position estimation of a mobile device. By focusing on an access points (AP) probabilistic existence, we develop an AP distribution map and a new position estimating framework. The developed approach can significantly enhance a radio fingerprint-based Wi-Fi Positioning System, especially in terms of the algorithms and data management. In comparison with existing approaches of fingerprint pattern matching algorithms, we achieve an improved performance in terms of both the average and deviation of the estimation error. All fingerprint data used in the developed test-bed are harvested from actual radio fingerprint measurements taken throughout Seoul, Korea. This demonstrates the practical usefulness of the proposed methodology.