Citation

Abdul Aziz, Nurul Ain and Ngo, Hong Yeap and Jabidin, Hafiz Zulhazmi and Choo, Kan Yeep and Tee, Connie and Muniandy, Sithi Vinayakam and Ibrahim@Ghazali, Siti Azlida and Abdul Rashid, Hairul Azhar and Mokhtar, Mohd Ridzuan and Zan, Mohd Saiful Dzulkefly (2024) Smart Optical Fiber Network Health Monitoring Using Deep Learning-Based Distributed Acoustic Sensing Technique. In: 2024 Multimedia University Engineering Conference (MECON), 23-25 July 2024, Cyberjaya, Malaysia.

Text Smart Optical Fiber Network Health Monitoring Using Deep Learning-Based Distributed Acoustic Sensing Technique.pdf - Published Version Restricted to Repository staff only Download (458kB)

Abstract

The quality of services and internet reliability depend crucially on the health and connectivity of the optical fiber network between the service providers and their clients. Unfortunately, the services can easily be disrupted due to the unintentional damage caused by the construction activities executed near the fiber network. This work demonstrated an integration of distributed acoustic sensing and deep learning techniques for fiber network health monitoring and break prevention. One of the dark fibers in a fiber cable buried along a road connecting the server room to a hostel on a university campus is used as the vibration sensor. A distributed acoustic sensing interrogator unit measured the vibration signals caused by the construction hand tools, such as sledgehammer, hoe, and shovel. The vibration signals were first denoised, and then the Mel-Spectrograms were calculated to obtain the Mel Frequency Cepstral Coefficients (MFCCs) by applying the discrete cosine transform to the Mel-Spectrograms. The MFCCs were treated as the training features for a simple Long Short-Term Memory (LSTM) network. The average accuracy score is more excellent than 70%, indicating that the proposed method has great potential for fiber network health monitoring and break prevention.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Fiber optic sensor
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television
Divisions:	Faculty of Engineering (FOE)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	12 Feb 2025 05:52
Last Modified:	12 Feb 2025 05:52
URII:	http://shdl.mmu.edu.my/id/eprint/13435

Downloads

Edit (login required)