The field of optical network management is witnessing a significant shift towards the adoption of machine learning (ML) techniques to improve network performance and reliability. Researchers are exploring the use of ML-based frameworks to model and predict optical power spectrum evolution, identify interference sources, and optimize amplifier gain. These innovative approaches aim to reduce the need for extensive data collection and training, making them suitable for deployment in real-world networks. Notably, the use of transfer learning and semi-supervised learning techniques is enabling the development of more accurate and scalable models. The application of these techniques is expected to have a significant impact on the management of multi-user optical networks, enabling more efficient detection and mitigation of interference and improved overall network performance. Noteworthy papers include:
- A novel ML-based attention framework for multi-span optical power spectrum prediction.
- A semi-supervised approach using internal amplifier features for EDFA gain modeling, which achieves high accuracy with minimal data requirement.