The Indoor Location Algorithm Based On RFID Technology Research

In the era of the Internet of things, many kinds of wireless technology have created the rapid development and the Radio Frequency Identification technology (Radio Frequency Identification, RFID) as the stars of indoor positioning, attracts the attention of the researchers, because of its non-contact, short time delay, the advantages of high precision and low cost, it has good adaptability for the complex environment indoor positioning. In this paper, Improvement of indoor location algorithm based on RFID,mainly done the following work:(1) After reading a large number materials of literature and books, and detailed analysis of the current research on indoor positioning technology at home and abroad and the development of the RFID technology, and sums up the development of RFID indoor positioning technology and deficiency.(2) This paper detailed introduces the current several kinds of wireless location technology which is mainly used in indoor positioning, including Wi-Fi location technology, ZigBee location technology, ultrasonic technology and the orientation of the RFID technology and the UWB technology, analyzes their advantages and disadvantages;In view of the existing RFID indoor localization algorithm, mainly for K neighbor algorithm, indoor localization algorithm based on least squares estimattion,indoor location algorithm based on bayesian filter theory and LANDMARC indoor positioning algorithms research, introduces their principle in detail, implementation methods and deficiencys.(3) For RFID indoor localization algorithm, this topic proposed to improve the indoor localization algorithm, using BP neural network model to fitt the RFID indoor loss, that is the mapping relationship between the distance d of the tag and the reader and the RSSI signal strength, and then use least-square method to estimate the backlog site location;According to deeply study of the BP neural network ,mind evolutionary algorithm is proposed to optimize the BP neural network's initial weights and threshold, this method improves the learning efficiency and convergence speed of neural network, but also can effectively avoids the neural network training process falling into a state of minimal value,improves the RFID indoor positioning accuracy.(4) The mind evolutionary algorithm, BP neural network and least-square method used in this article is implemented in MATLAB software, the software has the very strong function of data processing for the matrix, and the program runs faster, also gets the data and images from the programs, and has trong practicability.Through the simulation analysis, it is concluded that mind evolutionary algorithm to optimize BP neural network model, can significantly improve the performance of indoor location.Finally, summarized the research work of this article, and points out the next research content..