Machine Learning Driven Future Optical Networks

Recent advancements in machine learning have significantly enhanced the capability and flexibility of optical networks, paving the way for practical deployments. Shuangyi Yan’s research at the University of Bristol has been actively engaged in key areas such as future network architecture, Quality of Transmission (QoT) prediction, link behavior analysis, and dynamic routing solutions for optical networks.

In QoT prediction, Dr. Yan’s work focuses on developing scalable, network-level solutions, evolving from point-to-point to multi-link predictions and from single-channel to multi-channel predictions. For link behavior analysis, he has created tools for detecting link anomalies by analyzing time-series historical data. The dynamic routing solution, based on deep reinforcement learning algorithms, addresses the challenges of both multi-band and multi-dimensional optical networks.

In this webinar hosted by the Optics in Digital Systems Technical Group, Dr. Yan will present some of the latest experimental results in applying machine learning to optical networks. Additionally, he will discuss the essential technologies for practical deployment, including a unified monitoring and telemetry platform, digital twin, and an AI engine to manage all machine learning algorithms. Overall, the webinar aims to showcase the transformative potential of machine learning in revolutionizing optical networks and delivering practical benefits for future implementations.

Subject Matter Level: Intermediate - Assumes basic knowledge of the topic

What You Will Learn:
• Deep Reinforcement Learning for Resource provision and routing, modulation, spectrum allocation
• Digital twin and AI engine
• End-to-end QoT estimation
• Unified monitoring and telemetry platform
• Further leverage towards multi-dimensional networks

Who Should Attend:
• Graduate students
• PhD research scholars
• Easy stage researchers

About the Presenter: Shuangyi Yan from University of Bristol

Dr. Shuangyi Yan is a Senior Lecturer in the Smart Internet Lab at the University of Bristol. He received a B.E degree from Tianjin University in 2004. In 2009, he got a PhD degree from the Xi'an Institute of Optics and Precision Mechanics, CAS, China. From 2011 to 2013, he worked at the Hong Kong Polytechnic University. In May 2013, he joined the University of Bristol. His research focuses on machine-learning enabled dynamic optical networks and 5G Beyond networks, MCF-based multi-dimensional SDM networks, and data centre networks. He is the author or co-author of over 100 refereed publications, including several high-impact post-deadline papers . Dr Yan has delivered over a plentary of invited talks in conferences, such as OFC 2019, ONDM 2019 & 2018), ECOC 2018, and ACP 2017, ACP 2020. He also served as TPC members and Co-Chairs in several conferences, such as ECOC, EuCNC, ONDM, OECC, and ACP. Dr Yan is actively leading/participate several EU research projects and UK projects.