Optical Network AI and Digital Twins

Optical Network AI and Digital Twins

Project Overview:

Data-driven approaches and machine learning (ML) methods have garnered attention as potential solutions for managing the increased control complexity in fully disaggregated systems. Enhanced data collection and data-driven methods, designed to utilize such data, are crucial for making advancements in these disaggregated systems. As such, addressing the challenges of fully disaggregated systems necessitates experimental platforms for examining the interactions between new control and management systems and the physical transmission effects. 

Traditional optical networking experiments involved a small number (typically three to six) of nodes over short distances (e.g., a single transmission span). However, modern optical networks can scale to include hundreds or thousands of nodes, supporting signal transmission across dozens of nodes. Hence, new methods are imperative for experimentation to study the physical effects on a large scale and how they interact with innovative software controls, ML algorithms, or control hardware innovations. This shift towards data-driven controls for enhanced management and increased automation also necessitates new experimental emulation methods such as digital twins. 

A digital twin is a virtual model of a communication network that faithfully simulates the devices, communication links, operational conditions, and applications operating on the actual operational network. Through the creation of various settings within a controlled environment and the execution of multiple scenarios, digital twins offer an economical method to assess performance, anticipate the consequences of network alterations, enhance network administration, and facilitate informed decision-making. 

 

Funding Agency:

Science Foundation of Ireland https://www.sfi.ie/ 

 

Principal Investigator:

Dan Kilper

Research staff:  

Dr. Rishu Raj

Mr. Shuang Xie

 

Collaborators:

Duke University, NC, USA Duke University 

 

Location:

CONNECT SFI Research Centre, Trinity College  Dublin CONNECT | Ireland’s research centre for Future Networks and Communications (connectcentre.ie)