Machine Learning-Based Detection and Classification of Overlapping Fiber Anomalies
Authors:
- Gulmina MalikGulmina Malik
- Muhammad Umar MasoodMuhammad Umar Masood
- Ahtisham AliAAAhtisham Ali
- Mashboob Cheruvakkadu MohamedMMMashboob Cheruvakkadu Mohamed
- Stefano StraulluSSStefano Straullu
- Antonino NespolaANAntonino Nespola
- Sai Kishore BhyriSBSai Kishore Bhyri
- Antonio NapoliANAntonio Napoli
- Jo Ao PedroJPJo Ao Pedro
- Gabriele Maria GalimbertiGGGabriele Maria Galimberti
- Vittorio CurriVCVittorio Curri
- 26 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Technology Letters
- Vol. 38 (16) , 1157-1160
Abstract
This work presents a machine-learning framework for detecting and classifying independent and overlapping anomaly signatures in optical fiber infrastructure using state-of-polarization (SOP) dynamics. The framework exploits temporal variations of the Stokes parameters to characterize bending, tapping, shaking, and their simultaneous combinations. An experimentally collected dataset is used to evaluate multiple machine-learning models under realistic noise conditions representative of practical fiber monitoring environments. The results demonstrate that SOP-based features enable accurate identification of complex fiber disturbances, strengthening physical-layer monitoring and supporting the secure and reliable operation of next-generation optical networks.Keywords
Funding Information
- Project Piano Nazionale di Ripresa e Resilienza-Next Generation European Union (MUR–DM117/2023)
- EU’s Horizon Europe Research and Innovation Program [Agile Ultra Low Energy Secure Networks (ALLEGRO) Project] (101092766)
- EWOC (101073265)
This publication has 10 references indexed in Scilit:
- Intelligent Detection of Overlapping Fiber Anomalies in Optical Networks Using Machine Learning2025
- Resilient Anomaly Detection in Fiber-Optic Networks: A Machine Learning Framework for Multi-Threat Identification Using State-of-Polarization MonitoringAI, 2025
- Machine Learning for Predictive Multi-Event Detection in Fiber Optic Systems2025
- Fiber Eavesdropping Detection and Location in Optical Communication SystemPhotonics, 2025
- Vision Transformers for Anomaly Classification and Localization in Optical Networks Using SOP SpectrogramsJournal of Lightwave Technology, 2024
- Machine Learning-Based Polarization Signature Analysis for Detection and Categorization of Eavesdropping and Harmful Events2024
- Machine-learning-based anomaly detection in optical fiber monitoringJournal of Optical Communications and Networking, 2022
- Polarization sensing using submarine optical cablesOptica, 2021
- Evolution of Fiber-Optic Transmission and Networking toward the 5G EraiScience, 2019
- Experimental Study of Machine-Learning-Based Detection and Identification of Physical-Layer Attacks in Optical NetworksJournal of Lightwave Technology, 2019