Date of Publication :25th May 2024

Abstract: In these days, it is inevitable to apply machine learn- ing techniques in all fields of science and engineering including electromagnetic applications. Traditional antenna optimization methods involve solving modified versions of the original antenna design for each iteration, resulting in time-consuming processes dependent on convergence criteria and iteration times. This work introduces an approach for designing a planar antenna tailored for the Sub-6 GHz band of the fifth generation (5G) wireless network. The antenna has linear polarization characteristics and is intended to cover the 3.3-3.7 Ghz frequency band. The datasets for the suggested antenna are created using parametric analysis carried out in HFSS software to aid in the optimization process. Machine learning algorithms are subsequently employed to optimize the reflection coefficient (S11) of the antenna design. The results obtained through these machine learning techniques align closely with the simulated values in HFSS software. Finally, the optimized antenna design is fabri- cated and rigorously tested, offering a promising solution for efficient 5G communication within the sub-6 GHz spectrum.

Reference :