An overview of designing low-profile broadband high-gain microstrip patch antenna for 6G communication in FR3 band: ascension from traditional to machine learning approach

Abstract Antennas are designed and manufactured to stringent standards over time to meet the progressive demand of the market. The global telecom sector is now considering new mid-band spectrum allocations in the 7.125–24.25 GHz (FR3) bands due to the high bandwidth requirements for multi-Gbps communications. With the Internet of Everything (IOE), the role of miniaturized antennas became very much essential. This study surveys various miniaturized broadband, ultra-wideband, super-wideband, and band-notched antennas. This literature also analyzes various performance benchmark to assess achievement of the miniaturized broadband antenna through methodical tabulation. Notably, the analysis addresses the innate challenges in miniaturized broadband antenna design that necessitates the inclusion of gain improvement. Application of intelligent reflective surfaces such as frequency selective surface (FSS) for gain augmentation of various miniaturized broadband antennas is also explored in this study. Nowadays, machine learning (ML)-assisted optimization has shown promising results in antenna design. A significant number of studies addressing the optimization of the antenna and FSS using various evolutionary algorithms (EAs) have been explored. The survey finds that the inclusion of ML significantly improves the performance of the antenna, considering less simulation and computational time. Looking ahead, the survey explores various optimization algorithms to enhance the antenna performance parameters.