Convert Wi-Fi Signals for Fingerprint Localization Algorithm

This paper introduces a method which works together with the fingerprint localization algorithm in order to make it more applicable to many different mobile devices. The original fingerprint localization algorithm performs effectively only on the sample device used to build map in the training phase. Thus, full training is needed for each new device and this process is time-consuming (few hours or more). We propose an approach that can avoid this time-consuming process and only requires a quick calibration step (a few minutes). When tested with a fingerprint algorithm, our method achieves an accuracy of 69% on mobile devices. The experiment was performed on a 224-hour collected data from 16 different mobile devices.