An Artificial Neural Network Based Design of Triple-Band Microstrip Patch Antenna for WLAN Applications

With the development of communication system, multi-band antenna becomes more and more significant. In this paper, a triple-band microstrip patch antenna using artificial neural network techniques to optimize working bandwidth is proposed. The antenna is mainly composed of two metal patches and FR4 substrate. Three operating bands with the center frequency of 1.52 GHz (1.47 1.57 GHz), 2.46 GHz (2.43 2.48 GHz) and 2.79 GHz (2.78 2.81 GHz). Artificial neural network (ANN) model is trained and tested to optimize the bandwidth of the antenna, the simulation software HFSS is used to obtain 162 data sets of proposed antenna with four parameters related to patch dimension and substrate materials. According to the final test results, the average percentage error was within the acceptable range. The antenna also performs well in terms of gain. The antenna can be effectively used for WLAN applications.