Deep learning modeling approach for metasurfaces with high degrees of freedom
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement Metasurfaces have shown promising potentials in shaping optical wavefronts while remaining compact compared to bulky geometric optics devices. The design of meta-atoms, the fundamental building blocks of me...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
The Optical Society
2022
|
| Online Access: | https://hdl.handle.net/1721.1/142621 |
| _version_ | 1811090762076520448 |
|---|---|
| author | an, sensong Zheng, Bowen Shalaginov, Mikhail Y Tang, Hong Li, Hang Zhou, Li Ding, Jun Agarwal, Anuradha Murthy Rivero-Baleine, Clara Kang, Myungkoo Richardson, Kathleen A Gu, Tian Hu, Juejun Fowler, Clayton zhang, hualiang |
| author2 | Massachusetts Institute of Technology. Department of Materials Science and Engineering |
| author_facet | Massachusetts Institute of Technology. Department of Materials Science and Engineering an, sensong Zheng, Bowen Shalaginov, Mikhail Y Tang, Hong Li, Hang Zhou, Li Ding, Jun Agarwal, Anuradha Murthy Rivero-Baleine, Clara Kang, Myungkoo Richardson, Kathleen A Gu, Tian Hu, Juejun Fowler, Clayton zhang, hualiang |
| author_sort | an, sensong |
| collection | MIT |
| description | © 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement Metasurfaces have shown promising potentials in shaping optical wavefronts while remaining compact compared to bulky geometric optics devices. The design of meta-atoms, the fundamental building blocks of metasurfaces, typically relies on trial and error to achieve target electromagnetic responses. This process includes the characterization of an enormous amount of meta-atom designs with varying physical and geometric parameters, which demands huge computational resources. In this paper, a deep learning-based metasurface/meta-atom modeling approach is introduced to significantly reduce the characterization time while maintaining accuracy. Based on a convolutional neural network (CNN) structure, the proposed deep learning network is able to model meta-atoms with nearly freeform 2D patterns and different lattice sizes, material refractive indices and thicknesses. Moreover, the presented approach features the capability of predicting a meta-atom’s wide spectrum response in the timescale of milliseconds, attractive for applications necessitating fast on-demand design and optimization of a meta-atom/metasurface. |
| first_indexed | 2024-09-23T14:51:33Z |
| format | Article |
| id | mit-1721.1/142621 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T14:51:33Z |
| publishDate | 2022 |
| publisher | The Optical Society |
| record_format | dspace |
| spelling | mit-1721.1/1426212023-02-10T21:18:16Z Deep learning modeling approach for metasurfaces with high degrees of freedom an, sensong Zheng, Bowen Shalaginov, Mikhail Y Tang, Hong Li, Hang Zhou, Li Ding, Jun Agarwal, Anuradha Murthy Rivero-Baleine, Clara Kang, Myungkoo Richardson, Kathleen A Gu, Tian Hu, Juejun Fowler, Clayton zhang, hualiang Massachusetts Institute of Technology. Department of Materials Science and Engineering MIT Materials Research Laboratory © 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement Metasurfaces have shown promising potentials in shaping optical wavefronts while remaining compact compared to bulky geometric optics devices. The design of meta-atoms, the fundamental building blocks of metasurfaces, typically relies on trial and error to achieve target electromagnetic responses. This process includes the characterization of an enormous amount of meta-atom designs with varying physical and geometric parameters, which demands huge computational resources. In this paper, a deep learning-based metasurface/meta-atom modeling approach is introduced to significantly reduce the characterization time while maintaining accuracy. Based on a convolutional neural network (CNN) structure, the proposed deep learning network is able to model meta-atoms with nearly freeform 2D patterns and different lattice sizes, material refractive indices and thicknesses. Moreover, the presented approach features the capability of predicting a meta-atom’s wide spectrum response in the timescale of milliseconds, attractive for applications necessitating fast on-demand design and optimization of a meta-atom/metasurface. 2022-05-19T18:49:37Z 2022-05-19T18:49:37Z 2020 2022-05-19T18:37:16Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/142621 an, sensong, Zheng, Bowen, Shalaginov, Mikhail Y, Tang, Hong, Li, Hang et al. 2020. "Deep learning modeling approach for metasurfaces with high degrees of freedom." Optics Express, 28 (21). en 10.1364/OE.401960 Optics Express Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf The Optical Society Optica Publishing Group |
| spellingShingle | an, sensong Zheng, Bowen Shalaginov, Mikhail Y Tang, Hong Li, Hang Zhou, Li Ding, Jun Agarwal, Anuradha Murthy Rivero-Baleine, Clara Kang, Myungkoo Richardson, Kathleen A Gu, Tian Hu, Juejun Fowler, Clayton zhang, hualiang Deep learning modeling approach for metasurfaces with high degrees of freedom |
| title | Deep learning modeling approach for metasurfaces with high degrees of freedom |
| title_full | Deep learning modeling approach for metasurfaces with high degrees of freedom |
| title_fullStr | Deep learning modeling approach for metasurfaces with high degrees of freedom |
| title_full_unstemmed | Deep learning modeling approach for metasurfaces with high degrees of freedom |
| title_short | Deep learning modeling approach for metasurfaces with high degrees of freedom |
| title_sort | deep learning modeling approach for metasurfaces with high degrees of freedom |
| url | https://hdl.handle.net/1721.1/142621 |
| work_keys_str_mv | AT ansensong deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT zhengbowen deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT shalaginovmikhaily deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT tanghong deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT lihang deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT zhouli deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT dingjun deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT agarwalanuradhamurthy deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT riverobaleineclara deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT kangmyungkoo deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT richardsonkathleena deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT gutian deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT hujuejun deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT fowlerclayton deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom AT zhanghualiang deeplearningmodelingapproachformetasurfaceswithhighdegreesoffreedom |