An effective method for RFID tag antenna optimization based on artifical neural network

Abstract

Evolutionary algorithms combined with artificial neural network (ANN) have been applied in RFID tag antenna optimization platform. An effective method for RFID tag antenna optimization by particle swarm optimization (PSO) algorithm or genetic algorithm (GA) combined with ANN is presented in this paper. ANN is used to establish the non-linear model of tag antenna which is shown to be as accurate as an electromagnetic simulator and can be used for constructing the fitness function of PSO and GA. The PSO and GA optimizers are developed and executed in C++. Finally, this optimization method is turned out to be much more efficient than any electromagnetic simulator optimization. In addition, the PSO optimization results show that it is faster than GA.