Fully Unsupervised Detection of Physical Contacts on Subsea Cables via State-of-Polarization Monitoring

2026-07-01

AI Summary

This study addresses the challenge of real-time detection of physical contacts—such as those caused by trawling vessels—on submarine cables in completely unlabeled scenarios. The authors propose a fully unsupervised Fast-Slow Deep Support Vector Data Description (DSVDD) detector that automatically identifies anomalous events by analyzing time-series signals of the state of polarization (SOP). Without requiring any event labels during training, the method successfully detects all five confirmed trawling contact events, ranking them among the top 13 out of 122,174 records. Furthermore, it uncovers multiple previously unknown contact events subsequently validated through cross-verification. This approach substantially enhances the practicality and accuracy of unsupervised monitoring for submarine cable security.

Abstract

We present a fully unsupervised Fast-Slow DSVDD detector for continuous State-of-Polarization monitoring on a deployed subsea cable. Trained without event labels, it ranks all five confirmed trawler contacts within the top 13 of 122,174 recordings and surfaces additional corroborated cable-contact events.

Problem

• unsupervised detection
• physical contacts
• subsea cables
• State-of-Polarization monitoring

Innovation

• unsupervised learning
• subsea cable monitoring
• State-of-Polarization
• anomaly detection
• DSVDD