U-slot Loaded Half-Circled Microstrip Patch Antenna Analysis Using XGBOOST Machine Learning Algorithm

Abstract

Half-circled U-slot loaded antenna is studied using the HFSS and Machine Learning (ML) algorithm. The proposed work is for predicting the resonance frequency of the U-slot loaded antenna by providing the dimensions of the antennas.The proposed antenna is designed for Wi-MAX application with operating frequency of 3.4 GHz.The HFSS tool is being used for designing and analyzing fractal antennas and generating the training data. Parametric analysis of the designed U-slot-loaded half-circled antenna is developed by altering the half-circle radius, length of the U-slot and width. The data set is then given to the XGBoost ML algorithm for training the model. The XGBoost contains remarkably high processing speed and contains features like parallelization, cache optimization, and out-of-core computation which makes the perfect algorithm for predicting the resonance frequencies.U-slot loaded half-circled antenna offers a substantial size reduction, a wide impedance bandwidth, and a uniform radiation pattern on all sides.