Research On Action Recognition Technology Based On RFID

Since the concept of the Internet of Things was put forward,the technologies and applications of the Internet of Things have developed rapidly,and more and more scholars have devoted themselves to their research.Among them,wireless sensing technology has achieved a lot of success in various innovative applications,such as indoor positioning,activity recognition and behavior monitoring.Radio Frequency Identification(RFID),as a type of wireless sensing technology,has become the focus of research on Io T sensing technology due to its low cost,high precision,strong expansion capability and good stability.However,RFID-based perception recognition technology has problems such as low security and scene limitations,which have seriously affected the development of wireless perception technology.Therefore,further research on wireless perception technology is needed to achieve more secure and accurate perception recognition.This paper mainly focuses on the research of RFID-based action recognition,and proposes an RFID-based family identity authentication system and an RFID-based aerial handwriting recognition system.First,the paper introduces the development of RFID technology and the current research focus at home and abroad.Secondly,several current mainstream perceptual recognition technologies are introduced and compared with RFID perceptual recognition technologies.Finally,the two RFID perception recognition systems we designed are explained in detail.In order to solve the security problems of the home Internet of Things,we first designed a home identity authentication system based on RFID.In this system,we preprocess and segment the collected phase signals to obtain phase data for identification.In the action recognition stage,Convolutional Neural Networks(CNN)is introduced to realize the identity verification of family members.The second system is an aerial handwriting recognition system based on RFID,which constructs an aerial handwriting system by simulating Morse code.The system uses two RFID tags to represent the two basic characters in the Morse code,and then collects the phase signal of the tag group.Taking into account the differences between users,we choose to use Dynamic Time Warping(DTW)algorithm to match the phase data of the action to improve the accuracy of recognition.At the end of this paper,a large number of experiments have been carried out.The results show that the two motion recognition systems designed have high accuracy,which verifies the feasibility of the two systems.