Computation of Resonant Frequency and Gain from Inset Fed Rectangular Shaped Microstrip Patch Antenna Using Deep Neural Network

Abstract

This paper presents a deep neural network to predict the design parameters of an inset-fed rectangular shaped microstrip patch antenna. A multilayer perceptron based deep neural network has been proposed to predict the resonant frequency and gain values of antenna. Using five hidden layers and mini-batch gradient descent model the presented network converged closely with 1.30% and 1.56% mean absolute percentage errors compared with simulated frequency and gain values respectively of validation dataset. Hence the designed model can be used to determine resonant frequency and gain of inset-fed rectangular shaped microstrip patch antenna with any given dimensions.