Designing and Validation of Micro Strip Patch Antenna Using Artificial Neural Networks

Abstract

Computational modules based on neural networks have received attention as a novel and
effective technique for RF, microwave modeling, and the design of the antenna. It is possible
to train neural networks to learn the behavior of active or passive elements or circuits. In this
paper, the Designing of a Microstrip Patch Antenna using an Artificial Neural Network (ANN)
is presented. The antenna dimensions and parameters are determined with the help of Computer
Simulation Technology (CST) to produce an effective ANN model. The antenna design is
modeled using the Levenberg Marquardt optimization method and feed-forward backpropagation
neural network. Microstrip patch is initially created with the help of High-
Frequency Structure Simulator (HFSS) software running at 2.4 GHz (ISM band). A 2*2
microstrip planar array with a 2.4GHz operating frequency is constructed. After developing
and simulating findings using HFSS simulation software using the Finite Difference Time
Domain (FTDT) method, several neural networks are tested and trained to obtain the best
optimal outcomes. Neural networks are used for this research work. A radial basis function
neural network (RBF NN) and a feed-forward backpropagation technique are used for
optimization, and the results are compared to find the best optimum solution.