Super-resolution deep learning network for finite-difference time-domain simulations
When selecting the spatial grid size in finite-difference time-domain (FDTD) simulations, it is necessary to balance numerical dispersion and computational resources. Coarser grids are less computationally intensive but decrease simulation accuracy, while finer grids can reduce numerical dispersion...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/182288 |
| _version_ | 1826126984554479616 |
|---|---|
| author | Li, Haolin Liu, Shuo Tan, Eng Leong |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Li, Haolin Liu, Shuo Tan, Eng Leong |
| author_sort | Li, Haolin |
| collection | NTU |
| description | When selecting the spatial grid size in finite-difference time-domain (FDTD) simulations, it is necessary to balance numerical dispersion and computational resources. Coarser grids are less computationally intensive but decrease simulation accuracy, while finer grids can reduce numerical dispersion and enhance accuracy at the cost of increased simulation time. To address this challenge, this letter introduces a super-resolution deep learning network (SR-FDTDNet), which aims to improve FDTD simulation methods. SR-FDTDNet seeks to uncover the implicit mapping between coarse-grid and fine-grid simulation fields, enabling the reconstruction of fine-grid simulations from coarse-grid FDTD simulations. It can capture local information in the spatial and temporal dimensions of simulation data and model long-term dependencies through parallel computing, effectively addressing the complexity of FDTD simulation data. Results from practical electromagnetic simulation scenarios demonstrate that SR-FDTDNet significantly improves simulation efficiency while maintaining high accuracy in reconstructing fine-grid simulation fields. |
| first_indexed | 2025-03-09T14:26:43Z |
| format | Journal Article |
| id | ntu-10356/182288 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-03-09T14:26:43Z |
| publishDate | 2025 |
| record_format | dspace |
| spelling | ntu-10356/1822882025-01-20T08:03:43Z Super-resolution deep learning network for finite-difference time-domain simulations Li, Haolin Liu, Shuo Tan, Eng Leong School of Electrical and Electronic Engineering Engineering Coarse-fine grid reconstruction Deep learning network When selecting the spatial grid size in finite-difference time-domain (FDTD) simulations, it is necessary to balance numerical dispersion and computational resources. Coarser grids are less computationally intensive but decrease simulation accuracy, while finer grids can reduce numerical dispersion and enhance accuracy at the cost of increased simulation time. To address this challenge, this letter introduces a super-resolution deep learning network (SR-FDTDNet), which aims to improve FDTD simulation methods. SR-FDTDNet seeks to uncover the implicit mapping between coarse-grid and fine-grid simulation fields, enabling the reconstruction of fine-grid simulations from coarse-grid FDTD simulations. It can capture local information in the spatial and temporal dimensions of simulation data and model long-term dependencies through parallel computing, effectively addressing the complexity of FDTD simulation data. Results from practical electromagnetic simulation scenarios demonstrate that SR-FDTDNet significantly improves simulation efficiency while maintaining high accuracy in reconstructing fine-grid simulation fields. Ministry of Education (MOE) This work was supported in part by the Ministry of Education, Singapore, under Academic Research Fund Tier 1 (RG49/21), and in part by the Tertiary Education Research Fund under Grant MOE2021-TRF-019. 2025-01-20T08:03:43Z 2025-01-20T08:03:43Z 2024 Journal Article Li, H., Liu, S. & Tan, E. L. (2024). Super-resolution deep learning network for finite-difference time-domain simulations. IEEE Antennas and Wireless Propagation Letters, 23(12), 4763-4767. https://dx.doi.org/10.1109/LAWP.2024.3470522 1536-1225 https://hdl.handle.net/10356/182288 10.1109/LAWP.2024.3470522 2-s2.0-85205497255 12 23 4763 4767 en RG49/21 MOE2021-TRF-019 IEEE Antennas and Wireless Propagation Letters © 2024 IEEE. All rights reserved. |
| spellingShingle | Engineering Coarse-fine grid reconstruction Deep learning network Li, Haolin Liu, Shuo Tan, Eng Leong Super-resolution deep learning network for finite-difference time-domain simulations |
| title | Super-resolution deep learning network for finite-difference time-domain simulations |
| title_full | Super-resolution deep learning network for finite-difference time-domain simulations |
| title_fullStr | Super-resolution deep learning network for finite-difference time-domain simulations |
| title_full_unstemmed | Super-resolution deep learning network for finite-difference time-domain simulations |
| title_short | Super-resolution deep learning network for finite-difference time-domain simulations |
| title_sort | super resolution deep learning network for finite difference time domain simulations |
| topic | Engineering Coarse-fine grid reconstruction Deep learning network |
| url | https://hdl.handle.net/10356/182288 |
| work_keys_str_mv | AT lihaolin superresolutiondeeplearningnetworkforfinitedifferencetimedomainsimulations AT liushuo superresolutiondeeplearningnetworkforfinitedifferencetimedomainsimulations AT tanengleong superresolutiondeeplearningnetworkforfinitedifferencetimedomainsimulations |