Research On The Simulation And Optimization Software System For RFID Tag Antenna Based On Artificial Neural Network

The design and product of RFID tag antennas are important parts of RFID technology which should comply with the frequency standard of different counties. And the tag antennas should work in different environment (metal, fluent) which means that the tags will show quite different performance and makes the design and product of tags much more different.Nowadays the tag antennas are mainly simulated and optimized by EM simulation, which is very accurate but very slow and difficult to detect the performance change of the antennas after every change in size and structure. So the performance analysis of the tags and the definition of the optimized size are all very difficult.Based on the above issues, a fast simulation and optimization software for RFID tag antenna is presented in this paper. First classify, train and model ten RFID tag antennas and obtain the Artificial Neural Network (ANN) model for every antenna. And then using the VC6.0 platform in MFC interface and the MATLAB function, the database for antennas with different performance working in different environment and different frequency is established. So the user can fast locate the needed antenna. And the key parameters can be changed as need and the simulation result is available immediately. As the change of antenna chip impendence, the optimized sizes can be found immediately. All the work has greatly improved the efficiency of the antenna simulation and optimization, and meets the need of the research a for the tag antennas.The work was supported by "Research on RFID Tag Antenna Design Technology" of National High Technology Research Plan(863 plan).