Cognitive Network Automation: How Smart Can Optical Transport Networks Be?

Sunday, 06 June 05:00 – 07:30


Johannes K. Fischer, Fraunhofer Heinrich-Hertz-Institut​, Germany
Lida Galdino, Univ. College London​, UK
Paolo Monti, Chalmers University of Technology​, Sweden
Christine Tremblay, École de Technologie Supérieure, Montréal, Canada


5G and Beyond 5G services are expected to facilitate a transition towards smart communication infrastructures where network operations will account – in an automated fashion – for the specific end-user requirements and the network's status itself. 

This workshop aims at stimulating a debate over which challenges should be overcome in the metro/core transport segment and what advantages can be foreseen when enabling this transition.

On the one hand, it is expected that network automation will help supporting dynamic and real-time provisioning of services (e.g., via cognitive, intent-based, zero-touch operations), optimizing the usage of infrastructure capacity (e.g., reactive/proactive re-routing, defragmentation, combining photonic layer programmable infrastructure with advanced software applications), and lowering CAPEX and OPEX costs. All these benefits will be achieved by a combination of intent-based operations, Artificial Intelligence (AI) mechanisms, and Machine Learning (ML) tools supported by a highly flexible and software-driven data plane providing detailed physical-layer information the data analytics framework.

On the other hand, it can be expected that the benefits brought by network automation will come at a cost in terms of an increased complexity to guarantee: (i) service quality, (ii) secure and reliable operations, and (iii) consistency of operations across the various technological and administrative domains.


Jesse Simsarian, Nokia, USA
Network Automation: Progress and Pitfalls

Harald Bock, Infinera, Germany
A Path Towards a Smart Zero-touch Transport Network

Stephan Neidlinger, ADVA, Germany
Real-life Achievements and Gaps in Autonomous Optical Networks

Yvan Pointurier, Huawei, France
Network Automation – an Equipment Maker's Perspective

Stefan Melin, Telia Company, Sweden
Key Areas for Improvements in Optical Networks –How Automation Can Help Us As an Operator

David Côté, Ciena, Canada
Action Recommendation Engine: Using AI to Automate Network Operations

Daniel Kilper, Trinity College Dublin, Ireland
Smarter Hardware vs Smarter Software: Where is the Sweet Spot?

Takeshi Hoshida, Fujitsu, Japan
Advanced Optical Monitoring in DSP-based Receiver

Cristina Rottondi, Politecnico di Torino, Italy
ML-based QoT Estimation with Small Training Datasets and Under Measurement Uncertainty