Paper ID: 2409.03657

Unsupervised Anomaly Detection and Localization with Generative Adversarial Networks

Khouloud Abdelli, Matteo Lonardi, Jurgen Gripp, Samuel Olsson, Fabien Boitier, Patricia Layec

We propose a novel unsupervised anomaly detection approach using generative adversarial networks and SOP-derived spectrograms. Demonstrating remarkable efficacy, our method achieves over 97% accuracy on SOP datasets from both submarine and terrestrial fiber links, all achieved without the need for labelled data.

Submitted: Sep 5, 2024