Crowdsourcing-based Radio Map Update Automation for Wi-Fi Positioning Systems

Abstract

With the popularization of smartphones, the needs for indoor location information are rapidly growing these days. Wi-Fi based positioning technique has been widely used to provide positioning information indoors. In particular, fingerprint-based localization is preferred because of its advantage in accuracy. However, the accuracy of localization gradually degrades as the Wi-Fi environment changes. In order to prevent the accuracy degradation, a radio map, which stores the Wi-Fi environment information, should be updated to accommodate the changes. Recalibration is commonly used to update a radio map, but it usually requires considerable time and effort. In this paper, we propose a method that can update the radio map automatically by using fingerprints collected from numerous users. In order to tag the locations of the collected fingerprints more accurately, the data from various sensors such as accelerometer and gyroscope are used. The proposed method also uses optimization algorithms along with a filtering method to remove erroneous data. The evaluation results showed that the accuracy achieved by the method was comparable to that of manual calibration in spite of using user feedback data. This indicates that without recalibration, the fingerprint-based indoor positioning system can keep up its accuracy if it is used by many users.