Modelling and next-value prediction of beam propagation from grating structures using a simplified transformer model
In this study, a simplified transformer model is used to perform next-value prediction on light coupled out from silicon photonics gratings to free space. Finite-difference time-domain (FDTD) simulation is performed to simulate the electric field (E-field) in laser light coupled from gratings with p...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/179966 |
| _version_ | 1826122728204140544 |
|---|---|
| author | Lim, Yu Dian Tan, Chuan Seng |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Lim, Yu Dian Tan, Chuan Seng |
| author_sort | Lim, Yu Dian |
| collection | NTU |
| description | In this study, a simplified transformer model is used to perform next-value prediction on light coupled out from silicon photonics gratings to free space. Finite-difference time-domain (FDTD) simulation is performed to simulate the electric field (E-field) in laser light coupled from gratings with pitches of 0.6, 0.8, 1.0, 1.2, 1.4 and 1.6 µm, to free-space. Only E-field distribution from 0.6 µm is used in model training, and the trained transformer model is used to predict the E-field from the rest of the gratings. Prediction of accuracy up to 92.5% is obtained. The time taken for model training is 1908.4 seconds, which is significantly shorter than the conventional three-dimensional FDTD simulation that takes up to several hours. To further reduce the training time, transformer models can be trained with stepped datasets, but with compromised prediction accuracies. In summary, we demonstrated that the transformer model can be used to perform next-value E-field prediction using minimal training data. The developed and trained transformer model can be integrated to the state-of-the-art FDTD software to further expedite the existing FDTD simulation. |
| first_indexed | 2024-10-01T05:53:13Z |
| format | Journal Article |
| id | ntu-10356/179966 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:53:13Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1799662024-09-13T15:40:54Z Modelling and next-value prediction of beam propagation from grating structures using a simplified transformer model Lim, Yu Dian Tan, Chuan Seng School of Electrical and Electronic Engineering Institute of Microelectronics, A*STAR Engineering Electric field distributions Distribution transformers In this study, a simplified transformer model is used to perform next-value prediction on light coupled out from silicon photonics gratings to free space. Finite-difference time-domain (FDTD) simulation is performed to simulate the electric field (E-field) in laser light coupled from gratings with pitches of 0.6, 0.8, 1.0, 1.2, 1.4 and 1.6 µm, to free-space. Only E-field distribution from 0.6 µm is used in model training, and the trained transformer model is used to predict the E-field from the rest of the gratings. Prediction of accuracy up to 92.5% is obtained. The time taken for model training is 1908.4 seconds, which is significantly shorter than the conventional three-dimensional FDTD simulation that takes up to several hours. To further reduce the training time, transformer models can be trained with stepped datasets, but with compromised prediction accuracies. In summary, we demonstrated that the transformer model can be used to perform next-value E-field prediction using minimal training data. The developed and trained transformer model can be integrated to the state-of-the-art FDTD software to further expedite the existing FDTD simulation. Ministry of Education (MOE) Published version This work was supported by the Ministry of Education of Singapore AcRF Tier 2 (T2EP50121-0002 (MOE-000180-01)) and AcRF Tier 1 (RG135/23, RT3/23). 2024-09-09T00:59:43Z 2024-09-09T00:59:43Z 2024 Journal Article Lim, Y. D. & Tan, C. S. (2024). Modelling and next-value prediction of beam propagation from grating structures using a simplified transformer model. Optics Express, 32(18), 31533-. https://dx.doi.org/10.1364/OE.531050 1094-4087 https://hdl.handle.net/10356/179966 10.1364/OE.531050 18 32 31533 en T2EP50121-0002 (MOE-000180-01) RG135/23 RT3/23 Optics Express © 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement. application/pdf |
| spellingShingle | Engineering Electric field distributions Distribution transformers Lim, Yu Dian Tan, Chuan Seng Modelling and next-value prediction of beam propagation from grating structures using a simplified transformer model |
| title | Modelling and next-value prediction of beam propagation from grating structures using a simplified transformer model |
| title_full | Modelling and next-value prediction of beam propagation from grating structures using a simplified transformer model |
| title_fullStr | Modelling and next-value prediction of beam propagation from grating structures using a simplified transformer model |
| title_full_unstemmed | Modelling and next-value prediction of beam propagation from grating structures using a simplified transformer model |
| title_short | Modelling and next-value prediction of beam propagation from grating structures using a simplified transformer model |
| title_sort | modelling and next value prediction of beam propagation from grating structures using a simplified transformer model |
| topic | Engineering Electric field distributions Distribution transformers |
| url | https://hdl.handle.net/10356/179966 |
| work_keys_str_mv | AT limyudian modellingandnextvaluepredictionofbeampropagationfromgratingstructuresusingasimplifiedtransformermodel AT tanchuanseng modellingandnextvaluepredictionofbeampropagationfromgratingstructuresusingasimplifiedtransformermodel |