Related Experiment Video
Updated: Jun 25, 2025
05:30
Large Scale Energy Efficient Sensor Network Routing Using a Quantum Processor Unit
Published on: September 8, 2023
538
Environmental Surveillance through Machine Learning-Empowered Utilization of Optical Networks.
Hasan Awad1, Fehmida Usmani1,2, Emanuele Virgillito1
1Department of Electronics and Telecommunications, Polytechnic University of Turin, 10129 Turin, Italy.
Sensors (Basel, Switzerland)
|May 25, 2024
Summary
We developed a machine learning model using fiber optic networks for early earthquake detection. This system accurately identifies seismic waves, enabling faster earthquake warnings and improved public safety.
Area of Science:
- Geophysics
- Optical Engineering
- Machine Learning
Background:
- Existing seismic networks have limitations in coverage and speed.
- Terrestrial fiber optic infrastructure offers a potential solution for enhanced seismic monitoring.
Purpose of the Study:
- To develop and validate a machine learning model for early earthquake detection using optical mesh networks.
- To assess the system's capability for earthquake localization and early warning generation.
Main Methods:
- Simulated fiber optic strain data from ground displacement using a waveplate model.
- Trained and validated a machine learning model to detect primary seismic wave arrivals.
- Tested the model on an M4.3 earthquake using three interconnected optical mesh networks.
Main Results:
- Machine learning model achieved over 95% accuracy in detecting primary waves during validation.
- Real-world testing demonstrated 98% accuracy and a 1-second detection time.
- The system successfully localized an earthquake epicenter and calculated distances.
Conclusions:
- Interconnected optical mesh networks can effectively serve as a smart sensing grid for seismic monitoring.
- The developed machine learning approach enables rapid earthquake detection and localization.
- This technology offers significant potential for improving earthquake early warning systems.
