National Dark Fibre Facility

Lead Research Organisation: University College London
Department Name: Electronic and Electrical Engineering

Abstract

The National Dark Fibre Facility (NDFF) will provide the UK National Research Facility for dark fibre network research. A dark fibre network is a communications network, where it is possible to access and control the network at the optical layer, Layer1 (physical layer) in the seven layer Open Systems Interconnection (OSI) model of communications networks which underpins the internet.
NDFF will be a new, fully remotely configurable, flexible and high capacity research facility, building upon the success of the National Dark Fibre Infrastructure Service (NDFIS1) dark fibre network (2013-2018). This will allow UK universities and their industrial collaborators to develop and demonstrate future networks which require access to or control of the optical layer (OSI Layer1). NDFF will comprise:
1. A dark fibre network of scale sufficient for experiments representative of real-world applications (>600 km). Users will be able to connect their equipment directly to the installed fibres and control the optical layer, allowing experiments on new techniques, such as quantum encoded data, adaptive spectrum slicing and Software Defined Networks (SDN).
2. User experiment areas at multiple access nodes to interface directly with NDFF dark fibre. Interconnection for traffic generation and experiment control will also be possible UK-wide at Layer2 through services such as Janet Netpath.
3. Remotely programmable amplifiers, switches and dispersion compensation modules which will enable the transmission characteristics of the network to be varied and will allow dynamic configuration of the network topology by users.
4. Wavelength Selective Switches (WSS) to split optical channels into separate optical fibres (or merge them into one fibre). A Flexgrid WSS gives user defined channel widths, enabling research on new utilisation models for the optical spectrum, increasing available logical topologies and allowing concurrent experiments using different optical wavelengths.
5. A remotely configurable Layer2 and above network to enable research into dynamic and intelligent network management.
6. An SDN and Network Function Virtualisation (NFV) research platform for UK researchers, enabling them to upload network policies directly, monitoring and manipulating the optical properties of the network. UK researchers will be able to develop and test networks having optimised latency, traffic grooming, energy consumption or security properties.
7. A distributed processing infrastructure by linking sites that host servers, storage, memory and sensing. This will provide opportunities to study distributed Cloud and Fog infrastructures connected by high capacity reconfigurable optical networks. It will also provide nerve nodes that can perform network analytics offering users a new level of network programmability and adaptation.
8. The ability to test concepts in network security and resilience across all seven OSI model layers, something that is impossible with other networks. This is of particular importance as networks are starting to introduce software control and flexibility at Layer1, creating new security and resilience challenges for network control.
9. Training using dedicated research and technician support. Administration, user interface and dissemination will be the responsibility of a dedicated facility manager. In order to achieve the full potential of the facility, it is crucial to engage with the UK research community and promote the service. NDFF will engage in UK and international meetings and will bring together users at an annual user day. Web-based interfaces with users and potential users will be further developed.

Planned Impact

NDFF will have a great impact on the development of the internet, through (i) enabling advances in the core technical fields of Communications Engineering and Computer Science, via allowing new types of optical networks to be tested at scale whilst exposed to real world environmental effects and to interconnect research groups with a high capacity and flexible network and (ii) advancing all internet applications through the creation of a radically enhanced communication environment. Impacts would include:
a) Scientific/Academic
Large Data Sets:
1. Scientific: High volume data management and advanced visualisation for big Science (particle physics, radio-astronomy, biological sciences e.g. STFC SKA telescope). Remote optical sensing (e,g, Sensor Technology CDT, Cambridge, EP/L015889/1).
2. E-health: Very high quality transfer and visualisation of medical images for remote diagnosis and surgery (e.g. McKenna, Oxford, NS/A000024/1), remote monitoring for assisted living (NHS)
Low Latency:
1. Ubiquitous computing: New networked computing and storage architectures (Grid, Cloud and Fog) over high performance networks (e.g. Race, Lancaster, EP/R004935/1)
2. Media and entertainment: Ultra high definition broadcasting and digital cinema applications, distributed AR/VR, internet enabled gaming and e-sports (media companies, Cinegrid) - e.g. BBC R&D Multiplayer Broadcasting project
New Applications:
1. Quantum Technology: Distribution of quantum keys over networks, entanglement based quantum relays and repeaters to extend quantum links without trusted nodes. Demonstration of reliable quantum secured communication on shared fibres with conventional traffic. (UK Quantum Technology Programme, including InnovateUK project FQNet and Quantum Communication Hub Spiller, York EP/M013472/1).
2. New Physics: Ultrastable optical carrier transfer, enabling quantum technologies such as optical clocks, emerging applications such as relativistic geodesy, fundamental physics such as tests of General Relativity, optical carrier frequency transfer and dissemination, e.g. EURAMET ITOC project
3. Industrial/manufacturing applications such as networked control, autonomous vehicles and mega-city scale sensor networks (e.g Griffiths, Warwick, EP/N012380/1).
b) Economic
NDFF will enable research in support of large and growing markets in e.g. e-gaming ($23B), optical networks ($26B), optical sensing ($3B) and data centres ($35B) where over 20% of the entire EU spend is in the UK and is entirely dependent on high capacity resilient communications. Several company users use NDFIS1 for pre-competitive and collaborative research, or plan to do so (e.g. BT, Toshiba Research Europe, Microsoft, Oclaro, BBC). Concepts developed within the academic community can be trialled, aiding knowledge transfer, spin-out and standards activities. NDFF experiments would also benefit broadband development in the UK, contributing to the UK's economic and social development.
c) Skills and Training
NDFF will not only train PhD students and post-doctoral researchers but also engineers and apprentices in industry, both those directly associated with the delivery of NDFF, but also those who participate in field trials. This will provide a skilled workforce to meet the talent needs of both academia and industry.
d) Society
NDFF will enable users to contribute to the health and prosperity of UK society by trialling technologies, network concepts and applications in fields such as i) use of ICT in healthcare, ii) public access to data via high performance access to repositories, iii) secure networks enabling better protection of individual's personal and financial data. Users would be able to make use of the results of high profile trials using NDFF to influence policy and public opinion.

Publications

10 25 50