高动态室内环境下自适应定位研究-计算机技术专业论文.docxVIP

PAGE PAGE IV III III Abstract In the modern life of information, people’s demand on indoor positioning is increasing. Positioning application become more popular with mobile internet, but because of the diversity and variety of environment, it requirements the actual application ability and position precision of the system higher. In the last ten years, there exist many positioning system. Such as: GPS, GSM, infrared positioning, ultrasonic positioning, RFID, Bluetooth etc. But these systems have some limitations. At present all kinds of mobile terminal began to built-in wireless card, at the same time, the idea of wireless city makes environment appeared enough wireless access points, this all are beneficial for Wi-Fi positioning. The method of Wi-Fi includes transmission model and fingerprint. The transmission model is simple to calculate, but owe to indoor environment is complex, difficult to build accurately modeling, therefore positioning accuracy is low. Based on fingerprint, such as KNN, SVM, MR, BP etc. although can realize high precision, but not much think about the practical application, ignore the samples of utilization rate. Therefore, Indoor positioning face two challenges. First, variety of Wi-Fi environment, impact on the positioning accuracy direct. Secondly, although most of the fingerprint method, can greatly improve precision, but time-transference is very poor, namely after environmental changes, locating performance will be slowed sharply , it cant really put into practical application. Aiming at these problems, this paper puts forward a dynamic collaborative indoor locating method based on multiple terminals of OSCOELM. The method use Wi-Fi information which provided by terminals for co-location of dynamic, and update sample after locating. Experimental results show that the method can not only realize high precision positioning, but also can avoid repetitive calibration, beneficial to actual application. Particulary, without Wi-Fi or Wi-Fi