Threat Classification on Deployed Optical Networks Using MIMO Digital Fiber Sensing, Wavelets, and Machine Learning

TL;DR

This paper presents a machine learning-based method for classifying mechanical threats like jackhammers and excavators on optical networks using MIMO digital fiber sensing and wavelet transforms, achieving high accuracy in field conditions.

Contribution

It introduces a novel combination of MIMO digital fiber sensing, wavelet analysis, and transfer learning for real-time threat classification on deployed optical networks.

Findings

93% classification accuracy achieved

Effective detection of mechanical threats in field conditions

Utilizes wavelet transform for signal analysis

Abstract

We demonstrate mechanical threats classification including jackhammers and excavators, leveraging wavelet transform of MIMO-DFS output data across a 57-km operational network link. Our machine learning framework incorporates transfer learning and shows 93% classification accuracy from field data, with benefits for optical network supervision.