Indoor Localization Method And System Using Wi-Fi Fingerprinting

With the rapid development of wireless communication technology, there is a growing demand for wireless localization service. Traditional localization system like Global Positioning System(GPS) is widely used to solve the problem of localization in outdoor environments. However, unfortunately, it does not work well indoors because the signal from the GPS satellites is too weak to penetrate most buildings. Therefore, indoor localization issue has become an increasingly popular research topic in recent years. Among various wireless localization solutions, Wi-Fi fingerprint based localization technology has become one of the most promising due to the ubiquity of Wi-Fi network infrastructure and Wi-Fi capable devices.In this dissertation, an indoor localization approach is proposed which integrates Wi-Fi fingerprint based localization technology and inertial navigation system(INS). In this approach, two methods are proposed to maintain the fingerprint database, referred to as passive and active fingerprint database maintenance methods, respectively. Based on this approach, an indoor localization system is also developed on the Android platform. With the passive fingerprint database maintenance method, gait detection technology is first used to identify the user’s motion state and then only if the user is identified in a stationary state, respective Wi-Fi fingerprints are saved into the database. With the active fingerprint database maintenance method, the users themselves can evaluate and correct the localization results, thus training and using the localization system at the same time. Particularly a clustering based error detection method is adopted to detect the erroneous input location data, which can effectively avoid the contamination of the fingerprint database. In order to avoid the cumulative error of dead reckoning method, the system records the user’s trajectory. Once the system detects user’s behavior of leaving the room, current location is corrected to be the coordinates of the entrance to the room. With an algorithm which integrates Wi-Fi fingerprint based localization technology and INS, the user obtains the final localization result.A series of experiments are carried out on Android platform in the real indoor environment. First, we obtain the parameters for DBSCAN clustering algorithm through experiments. Second, to alleviate the localization deviation caused by device diversity, we use signal strength differences when calculating the fingerprint similarity. Third, with the technology of gait detection, the accuracy of step counting can reach more than 95%. Compared to the pure Wi-Fi fingerprint based localization algorithm, the algorithm which integrates Wi-Fi fingerprint based localization technology and INS achieves higher localization and tracking accuracy. In summary, the experiment results show the superiority of our proposed approach and the feasibility of our system.