Fast fault localization based on deep learning in optical networks

L. Chen; Y. Li; S. Ma; Y. Jing; H. Zhou; X. Gu; Q. Zheng; F. Wang; Y. Zhao

doi:10.1117/12.2666426

4 April 2023 Fast fault localization based on deep learning in optical networks

L. Chen, Y. Li, S. Ma, Y. Jing, H. Zhou, X. Gu, Q. Zheng, F. Wang, Y. Zhao

Proceedings Volume 12617, Ninth Symposium on Novel Photoelectronic Detection Technology and Applications; 126174Z (2023) https://doi.org/10.1117/12.2666426
Event: 9th Symposium on Novel Photoelectronic Detection Technology and Applications (NDTA 2022), 2022, Hefei, China

Abstract

In optical networks, fast and accurate fault localization ensures the normal operation and reliable transmission of a large number of network services, which has an important research significance. With the growing scale and complexity of optical networks, optical network fault localization becomes more challenging. Using the effective feature extraction capability of deep learning, this paper proposes a fault localization method based on deep learning, which improves the performance of fault localization in optical networks. First, the principle of deep learning and how it is applied to fault localization tasks are analyzed. Furthermore, the data is preprocessed to meet the input requirements of the deep learning model. Finally, using the preprocessed data set to train and verify the deep learning models with different parameters, and determine the optimal model parameters. The simulation results show that compared with the existing fault localization algorithms, the duration of the proposed fault localization method is shorter, the fault localization accuracy is higher, the fault localization delay is between 0.30 and 0.40ms, and the accuracy rate reaches more than 95%.

Citation Download Citation

L. Chen, Y. Li, S. Ma, Y. Jing, H. Zhou, X. Gu, Q. Zheng, F. Wang, and Y. Zhao "Fast fault localization based on deep learning in optical networks", Proc. SPIE 12617, Ninth Symposium on Novel Photoelectronic Detection Technology and Applications, 126174Z (4 April 2023); https://doi.org/10.1117/12.2666426

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available