Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jul 16, 2026

Writing Bragg Gratings in Multicore Fibers
08:48

Writing Bragg Gratings in Multicore Fibers

Published on: April 20, 2016

A Fast Demodulation Algorithm for Fibre Bragg Grating Based on the TimeMixer-LightGBM Hybrid Learning Framework.

Hang Gao1, Yizhe Su1, Kai Qian1

  • 1School of Intelligent Systems Science and Engineering, Hubei Minzu University, Enshi 445000, China.

Sensors (Basel, Switzerland)
|July 15, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Optimal dose window, formulation efficacy, and molecular mechanisms of curcumin in MASLD animal models: A systematic review and three-level meta-analysis.

Journal of ethnopharmacology·2026
Same author

Profiles and Networks: A Person-Centered Analysis of Emotional Labor, Burnout, and Change Fatigue Among Chinese Oncology Nurses.

Journal of nursing management·2026
Same author

Extracellular Vesicles: Key Mediators of Bioactive Molecule Transport in Plant-Microbe Interactions.

Molecular plant pathology·2026
Same author

Ultrasensitive and broadband micro-ring based optomechanical ultrasound sensor for photoacoustic imaging.

Optics express·2026
Same author

Effect of a multimodal bundled intervention on quality of recovery after laparoscopic colorectal cancer surgery: a single-center, single-blind, pragmatic randomized controlled trial.

World journal of surgical oncology·2026
Same author

Smartphone integrated portable oxidase-like hydrogel kit for UO<sub>2</sub><sup>2+</sup> colorimetric analysis.

Analytical methods : advancing methods and applications·2026

A new hybrid learning framework, TimeMixer-LightGBM, improves Fibre Bragg Grating (FBG) demodulation for structural health monitoring. It achieves picometre accuracy with high speed and noise robustness, enabling real-time applications.

Area of Science:

  • Optoelectronics and Photonics
  • Machine Learning for Sensing
  • Structural Health Monitoring

Background:

  • Fibre Bragg Grating (FBG) demodulation is crucial for structural health monitoring but faces challenges from spectral distortion, noise, and limitations in deep learning algorithms for real-time performance.
  • Existing methods struggle to balance accuracy and speed, hindering the application of FBG sensing in large-scale dynamic networks.

Purpose of the Study:

  • To develop a novel hybrid learning framework for high-precision and real-time FBG demodulation.
  • To address spectral distortion, noise, and computational efficiency limitations in current FBG demodulation techniques.
  • To validate the framework's performance on synthetically distorted and overlapping FBG spectra.

Main Methods:

  • A hybrid framework, TimeMixer-LightGBM, combining a pure MLP-based TimeMixer for feature extraction and a LightGBM model for regression.
Keywords:
LightGBMTimeMixerfibre Bragg grating (FBG) demodulationhybrid learning framework

More Related Videos

Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping
09:43

Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping

Published on: March 20, 2017

Related Experiment Videos

Last Updated: Jul 16, 2026

Writing Bragg Gratings in Multicore Fibers
08:48

Writing Bragg Gratings in Multicore Fibers

Published on: April 20, 2016

Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping
09:43

Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping

Published on: March 20, 2017

  • TimeMixer efficiently extracts multi-scale spectral sequence features from FBG reflection spectra.
  • LightGBM performs fast regression for centre wavelength shift prediction.

    • Main Results:

    • Achieved picometre-level accuracy (RMSE = 2.128 pm) with an average processing time of 0.08 ms per spectrum, a 4.5x speedup over state-of-the-art deep learning models.
    • Demonstrated excellent noise robustness, maintaining an average absolute error of 1.5 pm.
    • Outperformed existing methods in demodulating double-peaked overlapping spectra under various overlap conditions, with sub-millisecond inference times.

    Conclusions:

    • The TimeMixer-LightGBM framework offers a novel, effective solution for real-time, high-precision FBG demodulation.
    • Pure MLP architectures are effective for spectral analysis, and the hybrid approach balances accuracy, speed, and scalability.
    • Provides a foundation for deploying advanced FBG demodulation on embedded edge devices for structural health monitoring.