RSSI-based complex environment indoor positioning system

• Main Authors: QUE,YUAN-JUN, 闕源均
• Other Authors: HUNG,DUNG-SHING
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/30257584564219043893

碩士 === 銘傳大學 === 電腦與通訊工程學系碩士班 === 105 === People spend 80% of their time in indoors, however, the global positioning system cannot be used in indoors. Currently, indoor positioning system has difficulty to give a 100% of positioning because of environmental influences including pedestrian, WiFi etc.. It becomes crucial to increase the positioning accuracy, especially on dynamic environment. In this study, a wireless network system consisted of XBEE sensors, arduino and raspberry pi is constructed as a new positioning system. The XBEE sensors are used to collect the RSSI signal strength from the wireless network. The positioning process is given by the off-line database construction and on-line positioning. Then the RSSI signal strength and a fingerprint method are used to calculate the position. In the off-line stage, the collected RSSI signal is analyzed and established to be extreme value, average value, distance and noise ratio as the database connected to the on-line positioning. Up to 90% of poisoning accuracy is obtained in a 3m x 3m and 4m x 4m static (without interference) environment. By adding obstacles in the static environment, the RSSI noise increases and obviously affect the positioning accuracy in the dynamic environment. By adding one more sensor in the center of a dynamic environment and updating the off-line database by using the noise parameter, the 90% positioning accuracy in the 3m x 3m dynamic environment is obtained. The 80% positioning accuracy in the 4m x 4m dynamic environment is also obtained in real time.