WIFI Indoor Positioning System Based On Android

With the popularity of wireless LANs and mobile devices, demand for location based service(LBS) is growing, especially in complex indoor environments such as shopping malls,convention centers, airports in which the users of the mobile terminal often urgently need to know where they stand. Currently outdoor positioning technology is relatively mature,global satellite positioning system has been able to satisfy the needs of the people located outdoors, but when people indoors, since the barrier walls and other factors, GPS signals can not penetrate, unable to meet the needs of indoor positioning. WIFI indoor positioning systems are received widespread attention because of the low cost and other advantages.Smart phones now have universal coverage, Android system in the smart phone operating system market has been dominated by a very large proportion. Based on these situations,this paper focuses on the fingerprint localization algorithm based on clustering, and the use of this algorithm is designed and implemented WIFI indoor positioning system based on Android. The main work include:(1)I studied on some current domestic and foreign WIFI indoor positioning research which introduced the WIFI technology and the wireless positioning technology and their application areas,and mainly researched the location fingerprint algorithm.(2)We learn more about position fingerprint localization algorithm based on clustering, and analysis present clustering algorithm which is based on Receive Signal Strength(RSS). We find that clustering results will have some singular points and outliers due to the influence of the external factors such as the layout of the access point, building structure and so on.Therefore this thesis proposes the double clustering algorithm based on Receive Signal Strength and the position of reference points to preprocess the fingerprint data. A satisfactory clustering result can be obtained.In the precise positioning stage, KNN algorithm and compression perception algorithm are used to estimate the user's location coordinates.Through field testing, the proposed position scheme that double cluster combined with compression sense algorithm has a better location performance.(3)Design and implement the WIFI indoor position system based on android platform.This system consists of two applications written in JAVA language,one is used to collect finger-print on off-line stage,the other is used to finish real-time positioning on on-line stage.We finished thefingerprint database acquisition and positioning test on the second floor of the science building of our school,and the results showed that this system is stable,the positioning accuracy is within 3.5 meters.The proposed position system can provide the basic guarantee on service based on indoor location such as indoor location information awareness, route planning, the acquisition of around service information and so on.