Efficient Design Method for THz Metasurface Patterns

The traditional approach to design metasurfaces uses a combination of trial-and-error and simulation, which leads to low design efficiency. In this paper, based on deep learning and multi-objective genetic algorithm, an inverse design model for terahertz random metasurfaces with bandwidths from 0 TH...

Full description

Saved in:
Bibliographic Details
Published in2022 IEEE MTT-S International Microwave Workshop Series on Advanced Materials and Processes for RF and THz Applications (IMWS-AMP) pp. 1 - 3
Main Authors Yan, Teng, Chun, Li, Shaochen, Li, Yuhua, Xiao, Ling, Jiang
Format Conference Proceeding
LanguageEnglish
Published IEEE 27.11.2022
Subjects
Computational modeling
Conferences
Deep learning
Design methodology
Metasurfaces
Predictive models
Radio frequency
Online AccessGet full text

Cover

More Information
Summary:The traditional approach to design metasurfaces uses a combination of trial-and-error and simulation, which leads to low design efficiency. In this paper, based on deep learning and multi-objective genetic algorithm, an inverse design model for terahertz random metasurfaces with bandwidths from 0 THz to 2 THz is developed. by comparing with the traditional design method, the speed is increased by 40,000 times and the design efficiency is largely improved. The method proposed in this paper provides a new way for the design of complex terahertz metasurfaces.
ISSN:2694-2992
DOI:10.1109/IMWS-AMP54652.2022.10107182