Our research aim is to transform the design of optical networks to enable next-generation applications and capabilities of the digital communication infrastructure on different time- and length-scales. We aim to create an adaptive intelligent optical network that is able to dynamically provide capacity where and when it is needed, forming the backbone of the next- generation digital infrastructure.
We will focus on the development of machine-learning enhanced intelligent transceivers, architectures and network topologies capable of adapting to capacity and delay needs of new applications and continuously growing data demands. The overall goals are to
- Develop new approaches to the design of network topologies.
- Create a framework to quantify the fundamental limits of the amount of data carried over a network.
- Design a new class of low complexity intelligent transceivers capable of adapting over different time and distance scales.
- Introduce machine learning into all areas of network, transceiver design and operations.
To maximise our research mission, TRANSNET is guided by four interrelated research themes (RTs), each supported by two key academics from across the partner universities:
RT1: Intelligent network architecture and topologies
Delivers self-optimised networks by creating new architectures, frameworks and designs
RT2: Evolving transceiver intelligence and infrastructure-tailored digital signal processing
Explores software and hardware solutions to create dynamic transceiver designs
RT3: Learning algorithms for optical networking in the nonlinear regime
Addresses key challenges surrounding the introduction of intelligence into optical networks
RT4: Implementation and intelligent optical networks demonstration
Provides a platform to understand how to use the finite optical network resources