A review on machine learning and deep learning for various antenna design applications.

Area of Science:

• Electrical Engineering and Computer Science
• Artificial Intelligence in Communications

Background:

• Next-generation wireless networks increasingly depend on machine learning (ML) and deep learning (DL).
• ML and DL offer enhanced coverage and spectrum efficiency compared to traditional ground-based systems.
• Antenna design is a key application area benefiting from ML/DL's computational power and data handling.

Purpose of the Study:

• To provide a detailed discussion on the application of ML and DL in antenna design.
• To introduce the fundamental concepts of ML and DL.
• To highlight the advantages of ML/DL in antenna design over conventional methods.

Main Methods:

• Review and discussion of ML and DL concepts.
• Exploration of ML/DL applications across diverse antenna types.
• Analysis of feasibility, computational efficiency, and simulation reduction.

Main Results:

• ML and DL are highly effective for optimizing antenna design solutions.
• Applications span millimeter wave, body-centric, terahertz, satellite, UAV, GPS, and textile antennas.
• Demonstrated feasibility improvements, design acceleration, and reduced simulation requirements.

Conclusions:

• ML and DL offer significant advantages for modern antenna design.
• These AI techniques provide satisfactory and efficient results for complex antenna applications.
• The study underscores the transformative potential of AI in communication engineering.