Optimization of Terahertz Antenna Slot Dimensions Using Artificial Neural Network

Authors:

Abstract
Traditional optimization approaches based purely on iterative electromagnetic simulations, such as those used in HFSS, are computationally expensive and time-consuming, particularly when exploring a vast design space. To solve this difficulty, our research focuses on improving the slot dimensions of a microstrip patch antenna to increase its performance in the 117-124 GHz frequency band. With an architecture including an input layer, hidden layers, and an output layer, an Artificial Neural Network model was constructed in Python. The HFSS-generated dataset was used to train the ANN after it had been normalized and divided into training, validation, and testing sets. The ANN allowed for quick and effective design space exploration by predicting antenna performance for different slot configurations once it had been trained. Later HFSS simulations showed notable gains in antenna performance and confirmed the ideal slot size that the artificial neuron network had anticipated.

This publication has 38 references indexed in Scilit: