Optimization of a Metasurface Antenna Composed of Dual T-shaped Antenna Elements Based On Machine Learning

In the antenna design process, traditional electromagnetic full-wave simulation requires too much time and computing resources. As an emerging technology, machine learning(ML) is applied in the field of antenna design, with good performance, and can solve the problems in the above-mentioned traditional methods. This paper proposes a metasurface antenna composed of dual T-shaped antenna elements, and uses deep learning methods to optimize its geometric parameters. The resonant frequency of the proposed antenna is 7.9GHz and 13GHz. The gain in the 13GHz can reach 16.58dBi .