Smart optical networks enabled by performance monitoring and machine learning
The two main objectives of the project are:
(1) investigating machine learning (ML) techniques for quality of transmission (QoT) estimation and prediction in the optical layer;
(2) exploring ML methods for anomaly detection and proactive failure management at the network level.
ML-based methods for QoT estimation and prediction at the physical layer will be explored. ML techniques such as long-term short-memory (LSTM) and neural networks will be investigated for QoT prediction as ways to discover patterns in time-series and achieve dynamic and autonomous system margin control.
This page is only available in english
- École de Technologie Supérieure (ÉTS) (Academic, Canada)
- National Research Council Canada (Public, Canada)