Research On Location Algorithm Based On RFID System

Radio frequency identification technology (RFID) is an automatic data collection andnon-contact technology via the transmission medium of space electromagnetic waves, and it is alsothe core technology of the Internet of Things. The development of RFID has brought greatconvenience to not only the individuals but also the whole society. Currently, this technology isquite versatile in a broad range of applications, like management, transportation, location, trackingand so on. Due to its “non-contact” and “non line of sight” features, RFID technology must have abroad prospect to future development. Object location is one of the applications of RFID technology.Although RFID location hasn’t been widely used and its performance need to be improved, but ithas already become a popular solution of indoor location for its characteristics of high sensitivityand strong anti-interference. This paper aims to further improve the location accuracy and dynamicadaptability of the classical location algorithms.First, this paper introduces the development status, prospect and basic principles of RFIDtechnology. After that, four types of location methods (TOA, TDOA, AOA and RSSI) have beenelaborated. Then this paper introduces two classical RSSI location methods: LANDMARC andVIRE. By principle analysis and MATLAB simulation, this paper summaries the locationperformance of the two location methods.After figuring out the advantages and disadvantages of the two algorithms, the paper proposesthree kinds of modified location methods: VIRE location method based on Lagrange interpolation,adaptive threshold location method based on VIRE system and LAN-VIRE location method. VIRElocation method based on Lagrange interpolation has been used Lagrange interpolation instead oflinear interpolation in the original VIRE system, and enhance the system location accuracy andreliability. The adaptive threshold location method improves the environmental adaptability of thelocation system by setting dynamic threshold. LAN-VIRE location method combines theadvantages of LANDMARC and VIRE. In this algorithm, virtual reference tags and a normalrandom variable which represents the shadow effect have been introduced, and the problem of poorlocation accuracy of the sensing area boundary has been solved completely. The simulations ofthese algorithms show that they improve the location accuracy and stability of the location system.