Optimizing Elliptical Probe-Fed and Edge-Fed Antennas for Radar Applications Using a Two-Stage ANN Model

This paper presents an integrated approach for optimizing the design of Elliptical Probe-Fed and Edge-Fed Antennas for radar applications, leveraging a two-stage artificial neural network (ANN) model. The antennas operates at the radar band with a resonant frequency of 24.125 GHz. A two-stage ANN model was employed to predict antenna dimensions based on <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$S_11$</tex> and gain data, followed by a refinement stage for improved performance estimation. The ANN models were trained using data collect from HFSS (High-Frequency Structure Simulator). The optimized antenna demonstrates improved return loss and gain, validating the efficacy of the proposed approach for radar sysstem.