☆ 3.8 Proceedings Paper

Machine Learning Approach for Online Monitoring of Quality of Transmission Performance Indicators in Optical Fiber Networks

2021 EUROPEAN CONFERENCE ON OPTICAL COMMUNICATION (ECOC) (2021)

Machine learning regression for QoT estimation of unestablished lightpaths

Memedhe Ibrahimi et al.

Summary: Estimating the quality of transmission (QoT) of a candidate lightpath prior to its establishment using machine learning methods is important for resource allocation in optical networks. This study investigates ML regression approaches to estimate the distribution of the GSNR of unestablished lightpaths, comparing the performance of three regression approaches with synthetic data on a realistic network topology. The proposed estimation approaches provide benefits for network operators in making informed decisions about lightpath deployment.

JOURNAL OF OPTICAL COMMUNICATIONS AND NETWORKING (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Enhancing Lightpath QoT Computation With Machine Learning in Partially Disaggregated Optical Networks

Andrea D'Amico et al.

Summary: Increasing traffic demands are driving network operators to adopt disaggregated and open networking solutions for better utilization of optical transmission capacity, enabling a SDN approach. A machine learning algorithm trained on different ASE-shaped spectral loads is used to predict the OSNR component of the GSNR, combined with a QoT-E in the L-PCE to demonstrate a reliable LP feasibility predictor.

IEEE OPEN JOURNAL OF THE COMMUNICATIONS SOCIETY (2021)

添加到收藏夹

Article Optics