Inverse Design of Multi-band Reflective Polarizing Metasurfaces Using Generative Machine Learning

TL;DR

This paper introduces a data-driven, generative machine learning approach to design multi-band reflective polarizing metasurfaces, enabling efficient exploration of the design space and fabrication of stable, high-performance polarizers for satellite communication.

Contribution

It presents a novel use of generative adversarial networks for inverse design of multi-band polarizers, surpassing traditional empirical methods.

Findings

Designed dual- and triple-band polarizers with stable performance up to 30° incident angles.

Validated the designs through measurements of a fabricated prototype.

Demonstrated the effectiveness of the ML-based approach in metasurface design.

Abstract

Electromagnetic linear-to-circular polarization converters with wide- and multi-band capabilities can simplify antenna systems where circular polarization is required. Multi-band solutions are attractive in satellite communication systems, which commonly have the additional requirement that the sense of polarization is reversed {between adjacent bands}. However, the design of these structures using conventional \textit{ad hoc} methods relies heavily on empirical methods. Here, we employ a data-driven approach integrated with a generative adversarial network to explore the design space of the polarizer meta-atom thoroughly. Dual-band and triple-band reflective polarizers with stable performance over incident angles up to and including $30^\circ$, {corresponding to typical reflector antenna system requirements}, are synthesized using the proposed method. The feasibility and performance of the designed polarizer is validated through measurements of a fabricated prototype.