Research On Anti-multipath Device-free Localization Technology In RFID Environment

With the development of Internet of Things technology,many intelligent perception systems are progressively approaching people’s daily lives.Device-free localization(DFL),as a technology that can achieve positioning without the target being equipped with any equipment and actively participating in localization process,has broad development prospects in medical monitoring,emergency rescue and other application fields.In recent years,the DFL system based on received signal strength(RSS)has received widespread attention from scholars at home and abroad due to the convenient acquisition and calculation of RSS signals.Radio tomography imaging(RTI)is a DFL technology that maps the RSS attenuation value of a wireless communication link to a gridded positioning monitoring area,and finally displays the positioning results in a grayscale image.It has the advantages of strong real-time performance and low computational complexity.However,RTI only uses line of sight(LOS)path information to analyze the channel behavior of RSS,ignoring the multipath information contained in the actual received signal strength,and cannot effectively resist multipath interference.In view of the above problems,this thesis firstly builds a scene of an antenna array,which proves the reliability of static reflection multipath applied to passive positioning.On this basis,this thesis proposes a new training-free multipath enhancement(TFME-RTI)method.TFME-RTI proves the reliability of static reflection multipath applied to passive positioning,and solves the position of virtual reader antenna by introducing known reflectors in space.Then,it establishes the transformation model of static reflection multipath applied to RTI algorithm.Finally,this thesis conducts multiple experiments in two indoor scenarios.The median error of the TFME-RTI method under one target is 0.33 m,and the median error under two targets is 0.42 m.The experimental results show that the TFME-RTI method has high positioning accuracy.The Model-based DFL systems need to establish an accurate mathematical model in advance,but it may be difficult to achieve in complex multipath scenarios.Fingerprintbased DFL methods can combat multipath and noise interference problems through offline location fingerprint database and online matching mechanism.However,with the increase of the number of targets,building a multi-target fingerprint database usually consumes a lot of labor costs.To solve this problem,this thesis proposes a fast bayesian matching pursuit based device-free localization(FBMP-DFL)method.Under the framework of compressed ensing(CS),FBMP-DFL establishes the signal model of device-free multitarget localization by using the single target fingerprint library.In the sparse reconstruction stage,the fast Bayesian matching pursuit algorithm is used to estimate the number and location of multiple targets.The experimental results show that the FBMPDFL method can more accurately identify the number and position of targets below four targets.