Research And Improvement Of Indoor Positioning Technology Based On RFID

In recent years,with the development of network,location-based information services are used more and more in shopping malls,ports,warehouses and other scenes.Radio frequency identification is a kind of wireless communication technology.The technology can realize information switching without contact by using radio frequency signal through alternating magnetic field,and automatic identification is realized through the achieved information.It shows good performance in indoor positioning,and has the advantages of multi object identification,long identification distance and high identification speed.It has gradually become an important way of indoor positioning.Therefore,the research and analysis on positioning system based on RFID technology is of great importance for the improvement and development of positioning technology in China.Firstly,this paper describes the theroy of the RFID system,the LANDMARC algorithm and VIRE algorithm based on Received Signal Strength Indication positioning technology are analysed.Through the analysis of the layout of reference tags,the calculation method of virtual tag RSSI value,the threshold value,the coordinate calculation of the tags to be located,the shortcomings of traditional VIRE positioning system are summarized.Secondly,two kinds of location algorithms including NVIRE and D-NVIRE are proposed.NVIRE algorithm is improved in two parts,includung label layout and virtual label signal intensity value calculation.Real reference labels and virtual reference labels are added in boundary area.The intensity value of virtual label signal is calculated by using piecewise Newton interpolation method to solve the problem that the interpolation results deviate from the actual situation due to high order interpolation.The D-NVIRE algorithm improves the selection of labels and the method of coordinate calculation.Dynamic threshold is used to filter tags,and the initial value of the threshold is improved to reduce the number of iterations in the process of tag filtering,so as to improve the positioning efficiency.In order to reduce the influence of complex indoor environment on the coordinate calculation process,two weight factors used in traditional Vire positioning method are deleted.After running through the numbers of set in every possible combination,the preliminary coordinate is calculated by using centroid algorithm.To reduce the positioning error and obtain more accurate positioning,Taylor series expansion method is used to iterate the initial coordinates.Finally,the improved method proposed in this paper is analyzed and verified by experimental simulation.The results show that compared with the traditional VIRE algorithm,the NVIRE algorithm can reduce the positioning error in the boundary and center positioning areas,and improve the positioning stability.Compared with Vire algorithm,D-NVIRE algorithm with dynamic threshold of deviation correction has lower average positioning error of position,higher positioning efficiency and better stability.