Unsupervised Anomaly Detection and Localization with Generative Adversarial Networks

TL;DR

This paper introduces an unsupervised anomaly detection method using GANs and spectrograms, achieving over 97% accuracy on SOP datasets without requiring labeled data.

Contribution

The paper presents a novel GAN-based approach for unsupervised anomaly detection and localization using spectrograms, demonstrating high accuracy without labeled data.

Findings

Achieves over 97% accuracy on SOP datasets

Effective in both submarine and terrestrial fiber links

Operates without labeled training data

Abstract

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.