Design And Implementation Of Indoor Positioning Technology Using Wireless Network

With the rapid development of wireless communication network and the popularity of all kinds of cellphone based APPs, the mobile communication has become popular in daily life. Although the positioning based on mobile devices has made unprecedented achievements, the indoor location system, which has huge commercial value, hasn’t accepted for smart mobile device users. Therefore, it is necessary to come up with an optimized indoor location algorithm, which is suitable for mobile clients.In order to make this android based indoor location system comes true, this paper firstly explored the indoor positioning technology, and then analyzed the existing sensor networks and positioning methods. According to the previous researches, a system depending on fingerprint database was proposed to confirm the indoor environment. The complexity of the indoor environment weakens the accuracy and location speed of the current positioning method. As this article focused on the autonomous positioning of mobile terminals, which require high calculation speed, the positioning algorithm and K-nearest neighbor algorithm were optimized. The location system can greatly improve the calculation speed by reducing computational complexity. Meanwhile, positioning precision doesn’t decrease. The specific tasks of the paper are mainly as follows:First of all, by comparing the existing sensor networks and location method,Wi-Fi points were selected s signal source. Then through recording large number of signal strength by fingerprint-based method and storing them in database, the location information was achieved by matching the RSSI with the date in database.Secondly, an application for android mobile devices was developed to collect strength of WLAN signals. The signal strength from different terminals, different directions of terminal antennas, different collection time were analyzed, and the indoor signal’s characters were derived. Then the strength of signals were calculated preliminarily, and then sent to dominate server.Finally, an improved algorithm was put forward by studying on all kinds of algorithms. At the end of the paper, the algorithms were validated by a lot of simulation and experiments.The study results in this paper will enrich the application of intelligent fire-alert system and are meaningful to the optimization of indoor positioning accuracy and real-time performance.