An Indoor Localization Method Based On Location Fingerprinting In WLAN

With the rapid development of mobile internet and the popularity of intelligent terminals, indoor positioning based on WLAN has become an attractive research topic in the field of location-based services in the recent years because of its advantages such as low cost, high positioning accuracy, wide range of use, and so on.In this paper, an indoor localization approach based on location fingerprinting in WLAN is presented. Analyzing and identifying the key factors to indoor positioning, some solutions are designed. Experimental results testify to a strong feasibility in comparison with conventional indoor positioning methods. The details are as follows:First, the key influential factors to indoor localization are identified, including the relationship between different occlusion states of human body and the distance of the body from the terminal under different conditions, and how WiFi signal strength varies with the distance. Meanwhile, other factors are determined, such as sampling interval, the number of the access point (AP) deployment, the access point placement, the number of sampled signal, and so on. Furthermore, some research on the preprocessing of RSS, and the RSS threshold selection when getting offline database information are carried out to make localization more accurate.Second, the study focuses on improving the accuracy and stability of indoor positioning. Considering being occluded by the user itself and signal fluctuation of each AP, the localization algorithm via an orientation sensor is developed. First, transform the data from the orientation sensor used to provide information about the terminal into actual range information, then identify the APs being filtered according to access point placement and range information, and finally use the rest of APs to establish location-fingerprint database and position online. The proposed algorithm combines the data filtering, the query on dynamically changing data and the optimal AP localization to improve the accuracy and stability of positioning. Experimental results show that the occlusion caused by the user can be solved to improve the accuracy of the system. When positioning, give a preference to three APs with smaller fluctuation to query the database, and then determine whether to add a new AP to position in accordance with the number of sampling points queried, which not only ensure that in the sampling point density region the number of sample points to query can be controlled, and reduce the calculation of the system, but also ensure that in the sampling point sparse region the sampling points can be queried, and improve the stability of the positioning system.Third, an indoor positioning system based on Android platform is designed and implemented, in which the server-based system structure and the proposed algorithm are adopted. Tests in real environment reveal that the approach achieves higher accuracy and better stability.