English

Weather-Adaptive Multi-Step Forecasting of State of Polarization Changes in Aerial Fibers Using Wavelet Neural Networks

Networking and Internet Architecture 2024-09-06 v1 Machine Learning

Abstract

We introduce a novel weather-adaptive approach for multi-step forecasting of multi-scale SOP changes in aerial fiber links. By harnessing the discrete wavelet transform and incorporating weather data, our approach improves forecasting accuracy by over 65% in RMSE and 63% in MAPE compared to baselines.

Keywords

Cite

@article{arxiv.2409.03663,
  title  = {Weather-Adaptive Multi-Step Forecasting of State of Polarization Changes in Aerial Fibers Using Wavelet Neural Networks},
  author = {Khouloud Abdelli and Matteo Lonardi and Jurgen Gripp and Samuel Olsson Fabien Boitier and Patricia Layec},
  journal= {arXiv preprint arXiv:2409.03663},
  year   = {2024}
}

Comments

ECOC 2024