Automatically Inverse Design of FSSs by Conditional Variational Autoencoder and Equivalent Circuit Models With Large Language Models
Recently, deep learning has attracted intensive attentions in electromagnetic society, especially for inverse problems. In this work, we propose two language-guided automatic design frameworks for frequency selective surfaces (FSSs), where both of them are assisted by large language models (LLMs). O...
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on antennas and propagation Jg. 73; H. 11; S. 8922 - 8932 |
|---|---|
| Hauptverfasser: | , , , , , |
| Format: | Journal Article |
| Sprache: | Englisch |
| Veröffentlicht: |
New York
IEEE
01.11.2025
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Schlagworte: | |
| ISSN: | 0018-926X, 1558-2221 |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Abstract | Recently, deep learning has attracted intensive attentions in electromagnetic society, especially for inverse problems. In this work, we propose two language-guided automatic design frameworks for frequency selective surfaces (FSSs), where both of them are assisted by large language models (LLMs). On the one hand, the LLM is used to extract key desired features from language, which is then followed with a conditional variational autoencoder (CVAE) to generate S-parameters from the extracted features. With the generative S-parameters, an inverse surrogate model is designed to predict the geometry sizes of FSSs followed by LLM to output designed specifications in language. On the other hand, an inverse-design framework is further introduced to build the mapping from equivalent-circuit model to geometry sizes of FSS, where the loss is further constrained by a forward surrogate to alleviate nonunique problems in inverse design. The proposed framework is validated by designing numerous FSSs including high-pass, low-pass, bandpass, and band-stop types, where a representative structure is further fabricated and tested to verify the proposed methods. |
|---|---|
| AbstractList | Recently, deep learning has attracted intensive attentions in electromagnetic society, especially for inverse problems. In this work, we propose two language-guided automatic design frameworks for frequency selective surfaces (FSSs), where both of them are assisted by large language models (LLMs). On the one hand, the LLM is used to extract key desired features from language, which is then followed with a conditional variational autoencoder (CVAE) to generate S-parameters from the extracted features. With the generative S-parameters, an inverse surrogate model is designed to predict the geometry sizes of FSSs followed by LLM to output designed specifications in language. On the other hand, an inverse-design framework is further introduced to build the mapping from equivalent-circuit model to geometry sizes of FSS, where the loss is further constrained by a forward surrogate to alleviate nonunique problems in inverse design. The proposed framework is validated by designing numerous FSSs including high-pass, low-pass, bandpass, and band-stop types, where a representative structure is further fabricated and tested to verify the proposed methods. |
| Author | Liu, Jianfa Wang, Qian Shen, Jiajun Qu, Changhao Wang, Bao Wei, Zhun |
| Author_xml | – sequence: 1 givenname: Jiajun surname: Shen fullname: Shen, Jiajun organization: Zhejiang Key Laboratory of Intelligent Electromagnetic Control and Advanced Electronic Integration, College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China – sequence: 2 givenname: Jianfa orcidid: 0009-0003-3955-3518 surname: Liu fullname: Liu, Jianfa organization: Zhejiang Key Laboratory of Intelligent Electromagnetic Control and Advanced Electronic Integration, College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China – sequence: 3 givenname: Changhao surname: Qu fullname: Qu, Changhao organization: Zhejiang Key Laboratory of Intelligent Electromagnetic Control and Advanced Electronic Integration, College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China – sequence: 4 givenname: Bao orcidid: 0000-0002-4103-8393 surname: Wang fullname: Wang, Bao organization: Aeronautical Science Key Laboratory for High Performance Electromagnetic Windows, AVIC Research Institute for Special Structures of Aeronautical Composites, Jinan, China – sequence: 5 givenname: Qian orcidid: 0000-0001-9132-5086 surname: Wang fullname: Wang, Qian organization: Aeronautical Science Key Laboratory for High Performance Electromagnetic Windows, AVIC Research Institute for Special Structures of Aeronautical Composites, Jinan, China – sequence: 6 givenname: Zhun orcidid: 0000-0002-8699-3749 surname: Wei fullname: Wei, Zhun email: eleweiz@zju.edu.cn organization: Zhejiang Key Laboratory of Intelligent Electromagnetic Control and Advanced Electronic Integration, College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China |
| BookMark | eNpFkM1LAzEQxYMo2Kp3Dx4Cnrcm2WQ3eyy1VaGiYFFvS5qd1JQ1aZNdoXf_cFNa8DIfvPcG5jdEp847QOiakhGlpLpbjF9HjDAxykVVlFVxggZUCJkxxugpGhBCZVax4vMcDWNcp5VLzgfod9x3_lt1Vqu23eEn9wMhAr6HaFcOe4Nnb28RL3d44l1jO-udavG7ClYd530enPYNBKxcg6fb3v6oFlyHJzbo3nb4OYltxB-2-8JzFVaQqlv1Kg0H6RKdGdVGuDr2C7SYTReTx2z-8vA0Gc8zzXjZZcIw4MCN0KKSjVEgdcNybZLIWcObpYCcSE2YTj4pWGOAsryQVJWqNDq_QLeHs5vgtz3Erl77PqQnYp2zouSSlqRMLnJw6eBjDGDqTbDfKuxqSuo96jqhrveo6yPqFLk5RCwA_NspZZwzkf8B49F-SQ |
| CODEN | IETPAK |
| Cites_doi | 10.1109/TAP.2014.2361892 10.1109/TAP.2022.3195553 10.1109/JAS.2023.123618 10.1109/TAP.2005.844439 10.1109/ICCEM60619.2024.10558879 10.1109/TMAG.2013.2282532 10.1109/ISAP57493.2023.10388653 10.1109/TEMC.2023.3256388 10.1109/ACCESS.2021.3086777 10.1109/TAP.2022.3184545 10.1109/TAP.2016.2565694 10.1109/LAWP.2024.3399262 10.23919/ACES-China52398.2021.9581800 10.1109/TAP.2004.835289 10.1002/0471723770 10.1109/TEMC.2023.3312114 10.1109/LAWP.2024.3456838 10.1109/LAWP.2024.3362745 10.1007/s11633-023-1469-x 10.1109/CSRSWTC60855.2023.10426963 10.1109/tmtt.2024.3479872 10.1109/LAWP.2019.2944936 10.1109/TMAG.2015.2474816 10.1109/LAWP.2017.2743114 10.1109/TAP.2024.3366650 10.1109/TAP.2014.2357427 10.1109/TAP.2005.844438 10.1109/TAP.2019.2924909 10.1109/LAD62341.2024.10691722 10.1109/TEMC.2024.3427415 10.1109/TAP.2020.3027514 10.1109/RSEMW58451.2023.10202036 10.1109/TMTT.2023.3235066 10.1109/TAP.2021.3060142 10.1109/LAWP.2024.3372529 10.1109/ICMMT45702.2019.8992280 10.1109/TAP.2021.3096207 10.1109/TEMC.2023.3239747 10.1109/TAP.2015.2513423 10.1109/ACCESS.2023.3300381 10.1109/BigData59044.2023.10386911 10.1109/TAP.2023.3339217 10.1109/LAWP.2024.3477934 10.1109/LAWP.2023.3276998 10.1016/j.iotcps.2023.04.003 10.1109/LAWP.2006.875900 10.1109/TAP.2024.3456902 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025 |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025 |
| DBID | 97E RIA RIE AAYXX CITATION 7SP 8FD L7M |
| DOI | 10.1109/TAP.2025.3596796 |
| DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Electronics & Communications Abstracts Technology Research Database Advanced Technologies Database with Aerospace |
| DatabaseTitle | CrossRef Technology Research Database Advanced Technologies Database with Aerospace Electronics & Communications Abstracts |
| DatabaseTitleList | Technology Research Database |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 1558-2221 |
| EndPage | 8932 |
| ExternalDocumentID | 10_1109_TAP_2025_3596796 11124425 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: National Key Research and Development Program of China grantid: 2022ZD0117802 funderid: 10.13039/501100012166 |
| GroupedDBID | -~X 0R~ 29I 4.4 5GY 5VS 6IK 85S 97E AAJGR AASAJ AAWTH ABAZT ABFSI ABQJQ ABVLG ACGFO ACGFS ACIWK ACKIV ACNCT AENEX AETIX AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 E.L EBS EJD F5P HZ~ H~9 IAAWW IBMZZ ICLAB IDIHD IFIPE IFJZH IPLJI JAVBF LAI M43 O9- OCL P2P RIA RIE RNS RXW TAE TAF TN5 VH1 VJK VOH AAYXX CITATION 7SP 8FD L7M |
| ID | FETCH-LOGICAL-c247t-5f2e4e4f5c598dfae8cd23cf24742d4db5e308c02c2e4852dfe123681a7a7fc3 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 0 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001606666800027&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0018-926X |
| IngestDate | Sat Nov 01 15:13:58 EDT 2025 Sat Nov 29 06:55:58 EST 2025 Wed Nov 19 08:27:09 EST 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 11 |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037 |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c247t-5f2e4e4f5c598dfae8cd23cf24742d4db5e308c02c2e4852dfe123681a7a7fc3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 |
| ORCID | 0000-0001-9132-5086 0009-0003-3955-3518 0000-0002-8699-3749 0000-0002-4103-8393 |
| PQID | 3267481707 |
| PQPubID | 85476 |
| PageCount | 11 |
| ParticipantIDs | ieee_primary_11124425 crossref_primary_10_1109_TAP_2025_3596796 proquest_journals_3267481707 |
| PublicationCentury | 2000 |
| PublicationDate | 2025-11-01 |
| PublicationDateYYYYMMDD | 2025-11-01 |
| PublicationDate_xml | – month: 11 year: 2025 text: 2025-11-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | New York |
| PublicationPlace_xml | – name: New York |
| PublicationTitle | IEEE transactions on antennas and propagation |
| PublicationTitleAbbrev | TAP |
| PublicationYear | 2025 |
| Publisher | IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Publisher_xml | – name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| References | ref13 ref12 ref15 ref14 ref53 ref52 ref11 ref10 ref17 ref16 ref19 ref18 ref50 Brown (ref44) 2020 ref46 ref45 ref42 ref41 ref43 ref8 ref7 ref9 ref4 ref3 ref6 ref5 ref40 Touvron (ref47) 2023 ref35 ref34 ref37 ref36 ref31 ref30 ref33 ref32 ref2 ref1 Lalbakhsh (ref26) ref39 ref38 Grattafiori (ref48) 2024 ref24 ref23 ref25 ref20 Hu (ref51) 2021 ref22 ref21 ref28 ref27 ref29 (ref49) 2024 |
| References_xml | – ident: ref11 doi: 10.1109/TAP.2014.2361892 – ident: ref42 doi: 10.1109/TAP.2022.3195553 – ident: ref39 doi: 10.1109/JAS.2023.123618 – ident: ref23 doi: 10.1109/TAP.2005.844439 – ident: ref2 doi: 10.1109/ICCEM60619.2024.10558879 – ident: ref12 doi: 10.1109/TMAG.2013.2282532 – ident: ref4 doi: 10.1109/ISAP57493.2023.10388653 – ident: ref31 doi: 10.1109/TEMC.2023.3256388 – ident: ref27 doi: 10.1109/ACCESS.2021.3086777 – ident: ref36 doi: 10.1109/TAP.2022.3184545 – ident: ref53 doi: 10.1109/TAP.2016.2565694 – ident: ref7 doi: 10.1109/LAWP.2024.3399262 – ident: ref15 doi: 10.23919/ACES-China52398.2021.9581800 – ident: ref24 doi: 10.1109/TAP.2004.835289 – ident: ref1 doi: 10.1002/0471723770 – ident: ref21 doi: 10.1109/TEMC.2023.3312114 – ident: ref34 doi: 10.1109/LAWP.2024.3456838 – ident: ref8 doi: 10.1109/LAWP.2024.3362745 – year: 2023 ident: ref47 article-title: Llama 2: Open foundation and fine-tuned chat models publication-title: arXiv:2307.09288 – ident: ref46 doi: 10.1007/s11633-023-1469-x – year: 2021 ident: ref51 article-title: LoRA: Low-rank adaptation of large language models publication-title: arXiv:2106.09685 – ident: ref14 doi: 10.1109/CSRSWTC60855.2023.10426963 – ident: ref41 doi: 10.1109/tmtt.2024.3479872 – ident: ref5 doi: 10.1109/LAWP.2019.2944936 – ident: ref13 doi: 10.1109/TMAG.2015.2474816 – start-page: 1 volume-title: Proc. Int. Symp. Antennas Propag. (ISAP) ident: ref26 article-title: Multi-objective particle swarm optimization for the realization of a low profile bandpass frequency selective surface – ident: ref20 doi: 10.1109/LAWP.2017.2743114 – ident: ref22 doi: 10.1109/TAP.2024.3366650 – ident: ref10 doi: 10.1109/TAP.2014.2357427 – year: 2020 ident: ref44 article-title: Language models are few-shot learners publication-title: arXiv:2005.14165 – ident: ref19 doi: 10.1109/TAP.2005.844438 – ident: ref3 doi: 10.1109/TAP.2019.2924909 – ident: ref40 doi: 10.1109/LAD62341.2024.10691722 – ident: ref32 doi: 10.1109/TEMC.2024.3427415 – ident: ref18 doi: 10.1109/TAP.2020.3027514 – ident: ref17 doi: 10.1109/RSEMW58451.2023.10202036 – ident: ref28 doi: 10.1109/TMTT.2023.3235066 – ident: ref43 doi: 10.1109/TAP.2021.3060142 – volume-title: Independent Analysis of AI Models and API Providers year: 2024 ident: ref49 – ident: ref9 doi: 10.1109/LAWP.2024.3372529 – ident: ref16 doi: 10.1109/ICMMT45702.2019.8992280 – ident: ref33 doi: 10.1109/TAP.2021.3096207 – ident: ref29 doi: 10.1109/TEMC.2023.3239747 – ident: ref52 doi: 10.1109/TAP.2015.2513423 – ident: ref38 doi: 10.1109/ACCESS.2023.3300381 – ident: ref50 doi: 10.1109/BigData59044.2023.10386911 – ident: ref30 doi: 10.1109/TAP.2023.3339217 – ident: ref6 doi: 10.1109/LAWP.2024.3477934 – ident: ref35 doi: 10.1109/LAWP.2023.3276998 – ident: ref45 doi: 10.1016/j.iotcps.2023.04.003 – ident: ref25 doi: 10.1109/LAWP.2006.875900 – year: 2024 ident: ref48 article-title: The llama 3 herd of models publication-title: arXiv:2407.21783 – ident: ref37 doi: 10.1109/TAP.2024.3456902 |
| SSID | ssj0014844 |
| Score | 2.4827435 |
| Snippet | Recently, deep learning has attracted intensive attentions in electromagnetic society, especially for inverse problems. In this work, we propose two... |
| SourceID | proquest crossref ieee |
| SourceType | Aggregation Database Index Database Publisher |
| StartPage | 8922 |
| SubjectTerms | Autoencoders Band-pass filters Bandwidth Conditional variational autoencoder (CVAE) Data mining Design Equivalent circuits Feature extraction frequency selective surface (FSS) Frequency selective surfaces Inverse design Inverse problems Language large language model (LLM) Large language models Load modeling Machine learning Parameters Scattering parameters Silicon Training |
| Title | Automatically Inverse Design of FSSs by Conditional Variational Autoencoder and Equivalent Circuit Models With Large Language Models |
| URI | https://ieeexplore.ieee.org/document/11124425 https://www.proquest.com/docview/3267481707 |
| Volume | 73 |
| WOSCitedRecordID | wos001606666800027&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 1558-2221 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014844 issn: 0018-926X databaseCode: RIE dateStart: 19630101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSyQxEA4qHtyD78XxRQ5ePLR2p5NJchxGBw8igoPOrUnnwTYM07v9ELz7w62ke3wge_DSBJI0IVWV1CP1FUJnqcw1S3IZDbXQEU14HoGWoKLcJCLWTikpXSg2we_uxGwm7_tk9ZALY60Nj8_shW-GWL4pdetdZZcgl3AbEbaKVjnnXbLWe8iACtpBLicgwWQ4W8YkY3k5Hd2DJUjYRcqk95t8uYNCUZVvJ3G4XiZbP1zYNtrs9Ug86gi_g1bsYhf9-oQuuIdeR21TBkRWNZ-_YI-oUdUWX4U3G7h0ePLwUOP8BY9LH7cOPkH8CLZz7x_Efr7HuTS2wmph8PW_tgDGhKXgcVHptmiwL6U2r_FT0fzBt_5ROXw7B2jftY-mk-vp-Cbqiy5EmlDeRMwRSy11TDMpjFNWaENS7aCTEkNNzmwaCx0TDeMEI8ZZD-AiEsUVdzr9jdYW5cIeIOwHcstBYwOTDw4KGStFHbBESq2QQzJA50sqZH87aI0smCSxzIBimadY1lNsgPb9rn-M6zd8gI6XdMt64asz0Eg59cCD_PA_047Qhv97l1N4jNaaqrUnaF0_N0VdnQa-egNtWsz3 |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Lb9QwEB5BQaIceBaxUMAHLhzSJo69to-rpasillWlrmBvkeOHGmm1oXkg9c4PZ-xkeQhx4JJE8lixPB57Hp5vAN7mqjQ8K1UyNdIkLBNlglqCTkqbydR4rZXysdiEWK3kZqMuxmT1mAvjnIuXz9xJ-IyxfFubPrjKTlEu8TSi_Dbc4fjOhnStn0EDJtkAupyhDNPpZh-VTNXpenaBtiDlJzlXwXPyxykUy6r8tRfHA2bx8D-H9ggejJokmQ2sfwy33O4J3P8NX_ApfJ_1XR0xWfV2e0MCpkbTOvI-3togtSeLy8uWlDdkXofIdfQKks9oPY8eQhL6B6RL6xqid5acXfcVLk0cCplXjemrjoRiatuWfKm6K7IM18rxObhAx6YjWC_O1vPzZCy7kBjKRJdwTx1zzHPDlbReO2kszY3HRkYtsyV3eSpNSg3SSU6tdwHCRWZaaOFN_gwOdvXOPQcSCIUTqLOh0YdbhUq1Zh4XRc6cVFM6gXd7LhRfB3CNIholqSqQY0XgWDFybAJHYdZ_0Y0TPoHjPd-KUfzaAnVSwQL0oHjxj25v4N75-tOyWH5YfXwJh-FPQ4bhMRx0Te9ewV3zrava5nVcYz8APunQPg |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Automatically+Inverse+Design+of+FSSs+by+Conditional+Variational+Autoencoder+and+Equivalent+Circuit+Models+With+Large+Language+Models&rft.jtitle=IEEE+transactions+on+antennas+and+propagation&rft.au=Shen%2C+Jiajun&rft.au=Liu%2C+Jianfa&rft.au=Qu%2C+Changhao&rft.au=Wang%2C+Bao&rft.date=2025-11-01&rft.pub=IEEE&rft.issn=0018-926X&rft.volume=73&rft.issue=11&rft.spage=8922&rft.epage=8932&rft_id=info:doi/10.1109%2FTAP.2025.3596796&rft.externalDocID=11124425 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0018-926X&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0018-926X&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0018-926X&client=summon |