Research For Wireless Indoor Location Based On Support Vector Regression In IEEE802.11Environment

Location awareness is critical to the success of ubiquitous and pervasive computing. And it is very important for all kinds of applications of pervasive computing to get accurate position information of the mobile devices/users, which makes these applications capable of providing accurate and timely services for the users based on their position.Nowadays, GPS dominates the location area, but it can’t work in indoor environments because its radio signal suffers from the influence of the buildings, walls, etc. Existing indoor positioning technology can’t meet the needs of ubiquitous computing applications in the aspects like the using cost, deploying scope and portability, which limits the widespread of the location-based services.A new802.11-based indoor positioning method using support vector regression (SVR) is presented to address those challenges mention above. This method consists of two stages, offline training stage and online location stage. The accurate position prediction model is achieved in the offline training stage by SVR. In the online location stage, the exact position is determined according to the received signal strength (RSS) of the mobile devices and prediction model constructed in the offline stage. Due to the complex indoor environment, wireless channel congestion, obstructions and limitation of node communication range, RSS is vulnerable and changeable. To address the above issues, corresponding data filtering rules obtained through statistical analysis are applied in offline training period to improve the quality of training sample, and thus improve the quality of prediction model. In the online location period, k-times continuous measurement is utilized to obtain the high quality input, the online received signal strength, which guarantee the consistency with the training samples and improve the position accuracy of mobile devices.Performance evaluation and comprehensive analysis are done through intensive experiments. And the results show that it has a higher positioning accuracy when compared with the probability positioning method and neutral network based positioning method, and the demand for the storage capacity and computing power of the mobile devices is also low.