Dimension Forecast in Microstrip Antenna for C/X Band by Artificial Neural Network

Abstract

In this study, it is aimed to design C pattern array microstrip antenna for C Band (4 to 8 GHz), X Band (8 to 12 GHz) and Ku Band (12 GHz to 18 GHz). The proposed geometry was fed by coaxial probe. Optimum antenna was designed with Artificial Neural Network (ANN). Inputs of the network are return loses and operating frequencies in C/X/Ku band. On the other hand, there are six outputs such as 2-D feed points and other design dimensions. The simulated and forecasted return losses, radiation pattern and gain results are compared with each other. All simulation results were obtained with High Frequency Structure Simulator (HFSS) software. Also, training and test process of ANN was implemented in MATLAB software. Additionally, these simulation and predicted results were analyzed comparatively. In particular, results of antenna design based on ANN is so closed to real design. This technique can be used in microstrip antenna manufacturing.