Characterization and Performance Measurement of Minimum Bandwidth Ratio UWB CP Antenna Using Machine Learning

Abstract

In this article, a novel compact ultrawideband (UWB) circularly polarized (CP) inverted L-shaped-hook monopole (ILSHM) antenna is presented. The proposed ILSHM antenna incorporates a defective ground loop plane with two rectangular slits introduced to the primary radiator. These modifications play a crucial role in controlling current distribution and widening both the impedance bandwidth (IBW) and axial ratio bandwidth (ARBW). The measured UWB IBW is from 2.243 to 20.653 GHz (160.82%), while the ARBW is less than 3 dB from 3.216 to 18.985 GHz (143.45%). As a result, a minimum bandwidth ratio (BR) $\text {BR}|_{(10/3)\text {dB}}$ of $\approx ~1.121$ is achieved, ensuring nearly stable performance with minimal variation across different bandwidths. Furthermore, machine learning (ML) techniques, such as artificial neural networks (ANNs), are employed to predict the optimal design parameters of the proposed antenna. This approach automates and optimizes the design process, enhancing both adaptability and reliability. The ANN model is trained to identify the best parameter set for optimizing IBW and ARBW, streamlining the process of achieving high-performance characteristics. This novel approach to the UWB CP ILSHM antenna features a simple, compact design, minimum BR, and nearly stable radiation patterns and is highly suitable for UWB wireless applications.