Scholarlink

<p>This work presents a transfer learning approach for the forward design of composite unit cells of metasurfaces and frequency-selective surfaces (FSS). As a specific application, we target reconfigurable intelligent surfaces (RIS) for wireless communication. We demonstrate that forward models for dual-dipole unit cells can be trained using significantly fewer data by reusing pre-trained models of simpler, single-dipole structures. Our architecture reduces the required training data by up to a factor of 25 while maintaining a mean squared error (MSE) on the order of <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(10^{-2}\)</EquationSource> </InlineEquation>. After establishing the forward model for the dual-dipole RIS, we use it in an inverse design framework to synthesize a composite 2-bit RIS unit cell operating at 26.5&#xa0;GHz. The resulting RIS provides phase modulation in a 270<InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(^\circ\)</EquationSource> </InlineEquation> range by tuning the reverse bias voltage of integrated varactor diodes. Numerical simulations confirm the validity of the proposed approach, establishing transfer learning as a data-efficient and practical method for the design of composite unit cell structures.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Transfer learning for data-efficient modeling of composite metasurface and FSS unit cells

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月8日

查看全文
  • Alexander Wolff,
  • Lukas Mueller,
  • Steffen Klingel,
  • Marco Rahm

摘要

This work presents a transfer learning approach for the forward design of composite unit cells of metasurfaces and frequency-selective surfaces (FSS). As a specific application, we target reconfigurable intelligent surfaces (RIS) for wireless communication. We demonstrate that forward models for dual-dipole unit cells can be trained using significantly fewer data by reusing pre-trained models of simpler, single-dipole structures. Our architecture reduces the required training data by up to a factor of 25 while maintaining a mean squared error (MSE) on the order of \(10^{-2}\) . After establishing the forward model for the dual-dipole RIS, we use it in an inverse design framework to synthesize a composite 2-bit RIS unit cell operating at 26.5 GHz. The resulting RIS provides phase modulation in a 270 \(^\circ\) range by tuning the reverse bias voltage of integrated varactor diodes. Numerical simulations confirm the validity of the proposed approach, establishing transfer learning as a data-efficient and practical method for the design of composite unit cell structures.