An Analysis of Railway Activity Using Distributed Optical Fiber Acoustic Sensing
Abstract
1. Introduction
2. Materials and Method
2.1. Distributed Fiber Optic Sensing
2.2. Field Trial Setup
2.2.1. Fosina
2.2.2. CEF
2.2.3. DxS Acquisition Parameters Used
3. Results
3.1. Data from the Acquisition Campaign
3.2. Animals
3.3. Machine Learning
3.3.1. Dataset
3.3.2. Model Architecture and Design
3.3.3. Results
4. Discussion and Conclusions
4.1. Discussion
4.2. Conclusions
Author Contributions
Funding
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
Abbreviations
CEF | Centre d’Essais Ferroviaires |
CCTV | Closed-Circuit TeleVision |
CNN | Convolutional Neural Network |
DAS | Distributed Acoustic Sensing |
DFOS | Distributed Fiber Optic Sensing |
DSS | Distributed Strain Sensing |
DTGS | Distributed Temperature Gradient Sensing |
DTS | Distributed Temperature Sensing |
DTT | Dynamic Test Track |
FBE | Frequency Band Energy |
FBG | Fiber Bragg Grating |
OFS | Optical Fiber Sensor |
References
- Li, Z.; Zhang, J.; Wang, M.; Zhong, Y.; Peng, F. Fiber distributed acoustic sensing using convolutional long short-term memory network: A field test on high-speed railway intrusion detection. Opt. Express 2020, 28, 2925–2938. [Google Scholar] [CrossRef]
- Zeng, Y.; Zhang, J.; Zhong, Y.; Deng, L.; Wang, M. STNet: A Time-Frequency Analysis-Based Intrusion Detection Network for Distributed Optical Fiber Acoustic Sensing Systems. Sensors 2024, 24, 1570. [Google Scholar] [CrossRef]
- Wang, S.; Liu, F.; Liu, B. Semi-supervised deep learning in high-speed railway track detection based on distributed fiber acoustic sensing. Sensors 2022, 22, 413. [Google Scholar] [CrossRef]
- Wiesmeyr, C.; Litzenberger, M.; Waser, M.; Papp, A.; Garn, H.; Neunteufel, G.; Döller, H. Real-time train tracking from distributed acoustic sensing data. Appl. Sci. 2020, 10, 448. [Google Scholar] [CrossRef]
- Kowarik, S.; Hussels, M.T.; Chruscicki, S.; Münzenberger, S.; Lämmerhirt, A.; Pohl, P.; Schubert, M. Fiber optic train monitoring with distributed acoustic sensing: Conventional and neural network data analysis. Sensors 2020, 20, 450. [Google Scholar] [CrossRef]
- Tang, W.; Peng, L.; Li, Z.; Liu, K.; Pan, Z.; Yuan, S.; Xie, L. Detection of railway wheel and rail diseases based on distributed optical fiber acoustic sensing. In Proceedings of the Advanced Sensor Systems and Applications XII, SPIE, Online, 5–12 December 2022; Volume 12321, pp. 161–167. [Google Scholar]
- Gan, W.; Li, S.; Li, Z.; Sun, L. Identification of ground intrusion in underground structures based on distributed structural vibration detected by ultra-weak FBG sensing technology. Sensors 2019, 19, 2160. [Google Scholar] [CrossRef]
- Yan, S.; Shang, Y.; Wang, C.; Zhao, W.; Ni, J. Mixed intrusion events recognition based on group convolutional neural networks in DAS system. IEEE Sens. J. 2021, 22, 678–684. [Google Scholar] [CrossRef]
- Zhang, S.; He, T.; Li, H.; Fan, C.; Yan, Z.; Sun, Q. Modified data augmentation integration method for robust intrusion events recognition with fiber optic das system. J. Light. Technol. 2023, 42, 453–462. [Google Scholar] [CrossRef]
- Fu, W.; Yi, D.; Huang, Z.; Huang, C.; Geng, Y.; Li, X. Multiple event recognition scheme using variational mode decomposition-based hybrid feature extraction in fiber optic DAS system. IEEE Sens. J. 2023, 23, 27316–27323. [Google Scholar] [CrossRef]
- He, T.; Sun, Q.; Zhang, S.; Li, H.; Yan, B.; Fan, C.; Yan, Z.; Liu, D. A dual-stage-recognition network for distributed optical fiber sensing perimeter security system. J. Light. Technol. 2022, 41, 4331–4340. [Google Scholar] [CrossRef]
- Thakur, R.K. Railways to Use AI-Enabled Distributed Acoustic Sensing to Prevent Elephants Being Hit by Trains. 2025. Available online: https://www.newindianexpress.com/nation/2025/May/15/railways-to-use-ai-enabled-distributed-acoustic-sensing-to-prevent-elephants-being-hit-by-trains (accessed on 12 January 2024).
- Velha, P.; Nannipieri, T.; Signorini, A.; Morosi, M.; Solazzi, M.; Barone, F.; Frisoli, A.; Ricciardi, L.; Eusepi, R.; Icardi, M.; et al. Monitoring large railways infrastructures using hybrid optical fibers sensor systems. IEEE Trans. Intell. Transp. Syst. 2019, 21, 5177–5188. [Google Scholar] [CrossRef]
- Sasy Chan, Y.W.; Wang, H.P.; Xiang, P. Optical fiber sensors for monitoring railway infrastructures: A review towards smart concept. Symmetry 2021, 13, 2251. [Google Scholar] [CrossRef]
- Sasi, D.; Philip, S.; David, R.; Swathi, J. A review on structural health monitoring of railroad track structures using fiber optic sensors. Mater. Today Proc. 2020, 33, 3787–3793. [Google Scholar] [CrossRef]
- Klug, F.; Lackner, S.; Lienhart, W. Monitoring of railway deformations using distributed fiber optic sensors. In Proceedings of the Joint International Symposium on Deformation Monitoring (JISDM), Wien, Austria, 30 March–1 April 2016; Volume 1. [Google Scholar]
- Hartog, A.H. An Introduction to Distributed Optical Fibre Sensors; CRC Press: Boca Raton, FL, USA, 2017. [Google Scholar]
- Bolognini, G.; Hartog, A. Raman-based fibre sensors: Trends and applications. Opt. Fiber Technol. 2013, 19, 678–688. [Google Scholar] [CrossRef]
- Motil, A.; Bergman, A.; Tur, M. State of the art of Brillouin fiber-optic distributed sensing. Opt. Laser Technol. 2016, 78, 81–103. [Google Scholar] [CrossRef]
- Fosina. Available online: https://fosina.fr/innovation (accessed on 12 January 2024).
- Wu, H.; Chen, J.; Liu, X.; Xiao, Y.; Wang, M.; Zheng, Y.; Rao, Y. One-dimensional CNN-based intelligent recognition of vibrations in pipeline monitoring with DAS. J. Light. Technol. 2019, 37, 4359–4366. [Google Scholar] [CrossRef]
Acquisition Parameters | |
---|---|
Sampling frequency (Hz) | 500 |
Pulse duration (ns) | 10 |
Gauge length (m) | 5.7 |
Output spatial sampling (m) | 1.43 |
Layer (Type) | Output Shape | Parameters Number |
---|---|---|
conv1d2 (Conv1D) | (None, 19, 64) | 832 |
conv1d3 (Conv1D) | (None, 8, 64) | 49,216 |
maxpooling1d1 (MaxPooling1D) | (None, 1, 64) | 0 |
batchnormalization1 (BatchNormalization) | (None, 1, 64) | 256 |
dropout1 (Dropout) | (None, 1, 64) | 0 |
flatten1 (Flatten) | (None, 64) | 0 |
dense3 (Dense) | (None, 64) | 4160 |
dense4 (Dense) | (None, 32) | 2080 |
dense5 (Dense) | (None, 6) | 198 |
Total params: 56,742 (221.65 KB), Trainable params: 56,614 (221.15 KB) |
Layer (Type) | Output Shape | Parameters Number |
---|---|---|
timedistributed7 | (None, 5, 2461, 64) | 2624 |
timedistributed8 | (None, 5, 123, 64) | 0 |
timedistributed9 | (None, 5, 123, 64) | 256 |
timedistributed10 | (None, 5, 123, 128) | 163,968 |
timedistributed11 | (None, 5, 12, 128) | 0 |
timedistributed12 | (None, 5, 12, 128) | 512 |
timedistributed13 | (None, 5, 1536) | 0 |
bidirectional1 | (None, 128) | 819,712 |
dropout2 | (None, 128) | 0 |
dense6 | (None, 64) | 8256 |
dense7 | (None, 32) | 2080 |
dense8 | (None, 6) | 198 |
Total params: 997,606 (3.81 MB), Trainable params: 997,222 (3.80 MB) |
Classification Report on Test Set | ||||
---|---|---|---|---|
Class Number | Precision | Recall | f1-Score | Support |
1 | 1.00 | 0.99 | 1.00 | 95,000 |
2 | 0.65 | 0.69 | 0.67 | 95,000 |
3 | 0.66 | 0.76 | 0.71 | 95,000 |
4 | 0.75 | 0.67 | 0.70 | 95,000 |
5 | 0.61 | 0.57 | 0.59 | 95,000 |
6 | 0.73 | 0.70 | 0.71 | 95,000 |
accuracy | 0.73 | 570,000 | ||
macro avg | 0.73 | 0.73 | 0.73 | 570,000 |
weighted avg | 0.73 | 0.73 | 0.73 | 570,000 |
Classification Report on Test Set | ||||
---|---|---|---|---|
Class Number | Precision | Recall | f1-Score | Support |
1 | 1.00 | 0.97 | 1.00 | 95,000 |
2 | 0.87 | 0.94 | 0.67 | 95,000 |
3 | 0.84 | 0.85 | 0.71 | 95,000 |
4 | 0.79 | 0.85 | 0.70 | 95,000 |
5 | 0.92 | 0.80 | 0.59 | 95,000 |
6 | 0.88 | 0.88 | 0.71 | 95,000 |
accuracy | 0.88 | 21,000 | ||
macro avg | 0.88 | 0.88 | 0.88 | 21,000 |
weighted avg | 0.88 | 0.88 | 0.88 | 21,000 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Du, T.L.; Hartog, A.; Hilton, G.; Didelet, R. An Analysis of Railway Activity Using Distributed Optical Fiber Acoustic Sensing. Sensors 2025, 25, 4180. https://doi.org/10.3390/s25134180
Du TL, Hartog A, Hilton G, Didelet R. An Analysis of Railway Activity Using Distributed Optical Fiber Acoustic Sensing. Sensors. 2025; 25(13):4180. https://doi.org/10.3390/s25134180
Chicago/Turabian StyleDu, Thurian Le, Arthur Hartog, Graeme Hilton, and Roman Didelet. 2025. "An Analysis of Railway Activity Using Distributed Optical Fiber Acoustic Sensing" Sensors 25, no. 13: 4180. https://doi.org/10.3390/s25134180
APA StyleDu, T. L., Hartog, A., Hilton, G., & Didelet, R. (2025). An Analysis of Railway Activity Using Distributed Optical Fiber Acoustic Sensing. Sensors, 25(13), 4180. https://doi.org/10.3390/s25134180