Study Of RFID/WiFi Indoor Fusion Positioning Methods

In recent years, the increasing demand for precise positioning, WiFi positioning technology and RFID positioning technology are the two main indoor positioning technologies, with high positioning accuracy, low power consumption, low cost, suited to indoor environments positioning. In the WiFi network, we use the RSS(Received Signal Strength) of AP node and adapt the fingerprint positioning technolgy to determine the location of target location. And the method of RFID positioning technology to determine the location of an object is generally through the received signal strength RSSI(Received Signal Strength Indication) value which is needed to be received.Since the RSSI value of the received signal contains both the direct wave component, there are also contributions to many non-direct wave component. Because single positioning system based on RSSI measurement technology for positioning method often leads to that the positioning accuracy can not fully meet the requirement. Therefore, it needs to propose the use of WiFi and RFID establish multi-sensor data fusion system applys indoor positioning technology in order to improve positioning accuracy. Department of Education- China Mobile Research Fund "solution based on the integration of the indoor location of a wireless sensor network" is the support of this thesis. the methods of WiFi technology and RFID technology for the integration of indoor positioning were studied which lay the foundation for multi-sensor data fusion in this paper.The main work of this paper is that the WiFi location system use the analysis of WiFi networks localization algorithm for position estimation and use the position estimation method based on RSSI. Bayesian algorithm and nearest neighbor algorithm are analyzed and the nearest neighbor algorithm is identified as a WiFi positioning estimation measurement.The RFID positioning technology is studied and we use active RFID tags as small signal transmitter communicates with the reader by transmitting electromagnetic signals. It needs to use K nearest neighbor algorithm to match positioning labels location with real coordinates based on the large-scale model of signal propagation. And it needs to select the top three in the RSSI value of location labels and use LANDMARC algorithm to improve positioning accuracy.On the basis of research WiFi and RFID indoor positioning technology, for a single sensor can not meet the higher positioning accuracy of this issue, it puts forward to use the federated Kalman filtering in the positioning system to fuse the data obtained by WiFi, RFID location algorithm. Then it feeds the data back to sub-filters in order to improve the overall accuracy based on the iterative calculation.