Research On RFID Tag Positioning Method Based On Artificial Intelligence And Near-field Antenna

In recent years,the Internet of Things(IOT)technology are developing rapidly,began to spread in various fields,in the Internet,another industrial revolution after the mobile network.Although the origin of RFID technology is early,which has been practically applied since the 1940 s,but in the key process of the development of the Internet of Things,RFID technology has become a technology with low cost,wide application environment and mature development.And it became important technical means and broad application prospects of “connecting things with things” and “connecting things with internet” occupy a place in the Internet of Things industry.And the development of the Internet of Things technology also puts forward new requirements for indoor positioning technology.At present,the more mature satellite positioning technology cannot meet the high-precision requirements and the indoor environment with complex environment.Therefore,for people and things in a small complex environment Positioning has become an important requirement for the development of Internet of Things technology,and indoor positioning methods based on RFID technology have received widespread attention due to their advantages of small size,low cost,and strong environmental adaptability.This article first introduces the working mode of the RFID system and the energy transmission principle of the tag antenna.By calculating the maximum transmission distance of the RFID system,the relationship between the transmission distance and the received signal strength is analyzed.Then introduced two RFID positioning methods based on echo signal strength(RSSI)positioning by reference tags: LANDMARC algorithm and VIRE algorithm and their simulation analysis;at the same time using k-nearest neighbor and neural network two artificial intelligence algorithms are the same Based on the echo signal strength,the tag antenna is positioned and simulated.In this paper,two near-field antennas are designed,which are used for actual positioning tests in file cabinets and ordinary indoor environments,respectively.The positioning effects of LANDMARC algorithm,k-nearest neighbor method and neural network method are tested in two different scenarios.Tests and comparisons show that in practical scenarios,k-nearest neighbor algorithm and neural network algorithm have better adaptability to more complex closed environments,and the positioning effect is more accurate than LANDMARC method.