Machine Learning Based on Antennas Modeling for 5G and 6G Communication Systems: A Systematic Review

Artificial intelligence (AI)-aided communications have gained significant traction in recent years due to the widespread application of machine learning (ML) and deep learning (DL) machines with algorithms to solve math problems in wireless communications. This study offers an overview of the use of ML models in antenna design and optimization. This incorporates DL on ML frameworks, categories, and structure to get practical and broad insights using ML techniques for high throughput, quick data analysis, and prediction. This article also comprehensively reviews recent research papers on antenna design via ML. This includes an analysis of several ML algorithms that have been applied to produce antenna parameters such as the reflection coefficient (S-parameters), efficiency and gain values, and radiation patterns of the antennas. However, the current antenna design's structure, variables, and external factors remain complex. In addition, the expense of time and processing resources is inescapable and unacceptable to most designers. ML-based antennas have been created to increase antenna modeling efficiency and accuracy to solve these challenges. Techniques for modeling data may be used to predict the performance of an antenna for a certain set of antenna factors of design. As a result, this study highlights the most sophisticated applied ML techniques that have been presented to increase antenna modeling efficiency and accuracy. The results demonstrate that AI, ML, and DL may minimize simulation needs, predict antenna behavior, and reduce time with high accuracy.