Practical and Accurate Evaluation of Numerical Aperture and Beam Quality Factor in Photonic Crystal Fibers by Mechanical Learning

This paper presents a convolutional neural network (CNN) model, enhanced with the convolutional block attention module (CBAM), designed to accurately predict the beam quality factor M<sup>2</sup>, and numerical aperture (NA) of photonic crystal fibers. The integration of CBAM significant...

Full description

Saved in:
Bibliographic Details
Main Authors: Mengda Wei, Meisong Liao, Liang Chen, Yinpeng Liu, Wen Hu, Lidong Wang, Dongyu He, Tianxing Wang, Shizi Yu, Weiqing Gao
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Photonics Journal
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10767412/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850053344955990016
author Mengda Wei
Meisong Liao
Liang Chen
Yinpeng Liu
Wen Hu
Lidong Wang
Dongyu He
Tianxing Wang
Shizi Yu
Weiqing Gao
author_facet Mengda Wei
Meisong Liao
Liang Chen
Yinpeng Liu
Wen Hu
Lidong Wang
Dongyu He
Tianxing Wang
Shizi Yu
Weiqing Gao
author_sort Mengda Wei
collection DOAJ
description This paper presents a convolutional neural network (CNN) model, enhanced with the convolutional block attention module (CBAM), designed to accurately predict the beam quality factor M<sup>2</sup>, and numerical aperture (NA) of photonic crystal fibers. The integration of CBAM significantly improves the model&#x0027;s feature extraction capability by enabling it to focus on key features and filter out irrelevant information. Simulation results demonstrate that the model achieves a mean relative error of only 0.381&#x0025; for M<sup>2</sup> and 2.293&#x0025; for NA, outperforming convolutional models without attention mechanisms. With a prediction time of approximately 7 ms, the model allows for rapid and efficient predictions of M<sup>2</sup> and NA. Moreover, when the noise factor remains below 0.32, the model&#x0027;s prediction error shows minimal fluctuation, highlighting its robustness. Comparative experimental analysis further validates the model&#x0027;s effectiveness. This approach offers a reliable and efficient solution for fast, accurate measurement of M&#x00B2; and NA, with significant implications for the prediction and analysis of beam performance in various applications.
format Article
id doaj-art-72591f76a7f34b2aaac2249ca3bba9ea
institution DOAJ
issn 1943-0655
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Photonics Journal
spelling doaj-art-72591f76a7f34b2aaac2249ca3bba9ea2025-08-20T02:52:34ZengIEEEIEEE Photonics Journal1943-06552025-01-011711810.1109/JPHOT.2024.350662210767412Practical and Accurate Evaluation of Numerical Aperture and Beam Quality Factor in Photonic Crystal Fibers by Mechanical LearningMengda Wei0https://orcid.org/0009-0002-7675-5592Meisong Liao1https://orcid.org/0009-0000-5713-6716Liang Chen2https://orcid.org/0000-0001-5941-3174Yinpeng Liu3https://orcid.org/0009-0006-5490-6125Wen Hu4Lidong Wang5https://orcid.org/0009-0007-2106-1952Dongyu He6https://orcid.org/0009-0008-3106-4345Tianxing Wang7Shizi Yu8https://orcid.org/0000-0002-2923-1393Weiqing Gao9https://orcid.org/0000-0001-6181-5965Advanced Laser and Optoelectronic Functional Materials Department, Special Glasses and Fibers Research Center, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai, ChinaAdvanced Laser and Optoelectronic Functional Materials Department, Special Glasses and Fibers Research Center, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai, ChinaYangtze Delta Region Academy of Beijing Institute of Technology, Jiaxing, ChinaAdvanced Laser and Optoelectronic Functional Materials Department, Special Glasses and Fibers Research Center, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai, ChinaAdvanced Laser and Optoelectronic Functional Materials Department, Special Glasses and Fibers Research Center, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai, ChinaAdvanced Laser and Optoelectronic Functional Materials Department, Special Glasses and Fibers Research Center, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai, ChinaAdvanced Laser and Optoelectronic Functional Materials Department, Special Glasses and Fibers Research Center, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai, ChinaAdvanced Laser and Optoelectronic Functional Materials Department, Special Glasses and Fibers Research Center, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai, ChinaAdvanced Laser and Optoelectronic Functional Materials Department, Special Glasses and Fibers Research Center, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai, ChinaSchool of Electronic Science and Applied Physics, Hefei University of Technology, Hefei, ChinaThis paper presents a convolutional neural network (CNN) model, enhanced with the convolutional block attention module (CBAM), designed to accurately predict the beam quality factor M<sup>2</sup>, and numerical aperture (NA) of photonic crystal fibers. The integration of CBAM significantly improves the model&#x0027;s feature extraction capability by enabling it to focus on key features and filter out irrelevant information. Simulation results demonstrate that the model achieves a mean relative error of only 0.381&#x0025; for M<sup>2</sup> and 2.293&#x0025; for NA, outperforming convolutional models without attention mechanisms. With a prediction time of approximately 7 ms, the model allows for rapid and efficient predictions of M<sup>2</sup> and NA. Moreover, when the noise factor remains below 0.32, the model&#x0027;s prediction error shows minimal fluctuation, highlighting its robustness. Comparative experimental analysis further validates the model&#x0027;s effectiveness. This approach offers a reliable and efficient solution for fast, accurate measurement of M&#x00B2; and NA, with significant implications for the prediction and analysis of beam performance in various applications.https://ieeexplore.ieee.org/document/10767412/Attention mechanismbeam quality factordeep learningnumerical aperture
spellingShingle Mengda Wei
Meisong Liao
Liang Chen
Yinpeng Liu
Wen Hu
Lidong Wang
Dongyu He
Tianxing Wang
Shizi Yu
Weiqing Gao
Practical and Accurate Evaluation of Numerical Aperture and Beam Quality Factor in Photonic Crystal Fibers by Mechanical Learning
IEEE Photonics Journal
Attention mechanism
beam quality factor
deep learning
numerical aperture
title Practical and Accurate Evaluation of Numerical Aperture and Beam Quality Factor in Photonic Crystal Fibers by Mechanical Learning
title_full Practical and Accurate Evaluation of Numerical Aperture and Beam Quality Factor in Photonic Crystal Fibers by Mechanical Learning
title_fullStr Practical and Accurate Evaluation of Numerical Aperture and Beam Quality Factor in Photonic Crystal Fibers by Mechanical Learning
title_full_unstemmed Practical and Accurate Evaluation of Numerical Aperture and Beam Quality Factor in Photonic Crystal Fibers by Mechanical Learning
title_short Practical and Accurate Evaluation of Numerical Aperture and Beam Quality Factor in Photonic Crystal Fibers by Mechanical Learning
title_sort practical and accurate evaluation of numerical aperture and beam quality factor in photonic crystal fibers by mechanical learning
topic Attention mechanism
beam quality factor
deep learning
numerical aperture
url https://ieeexplore.ieee.org/document/10767412/
work_keys_str_mv AT mengdawei practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning
AT meisongliao practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning
AT liangchen practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning
AT yinpengliu practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning
AT wenhu practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning
AT lidongwang practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning
AT dongyuhe practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning
AT tianxingwang practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning
AT shiziyu practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning
AT weiqinggao practicalandaccurateevaluationofnumericalapertureandbeamqualityfactorinphotoniccrystalfibersbymechanicallearning