The Optical Networks Group (ONG) research is on fundamental properties of light and how it can be used to generate, carry, route and process data for different applications.
The work of the ONG Group focuses on optical communications systems and networks, operating on all time and length scales. These networks underpin the global communications infrastructure and the Internet, and are increasingly being used inside data centres. This helps us understand what is needed to support the increasing broadband data demands of a modern, global society.
Through the study of future optical fibre network architectures and transmission physics our research and innovations will enable new applications, essential to our digital lives of today.
The ONG has close interactions with leading industry and research groups around the world with many coming to use our advanced, state-of-the art optical fibre transmission and network test-bed.
We are justifiably proud of our focus on researcher development and their successes!
Many of our alumni have won major prizes, including some 10 IEEE Postgraduate Fellowship Awards for best PhD students in photonics worldwide as well as a host of best paper and other accolades. Most are now leading industrial and university research around the world on next-generation optical networks.
Our current research
We are interested in all linear and nonlinear phenomena related to the propagation of optical signals in various media, including different types of fibres and free-space.
In particular, we work to maximise optical fibre network capacity, though a combination of new research in nonlinear physics, information theory, machine learning and digital signal processing.
However, the next-generation digital infrastructure needs more than raw capacity – it requires flexible resources and low delays, the ability to allocate capacity when and where it is needed. How to build such intelligent, flexible and secure networks is a major focus of our current research.
Our current research is focused on these directions:
- Adaptive optical networks
- High-speed, ultrawideband optical fibre transmission systems
- Intelligent Transceivers
This research is enabled by a range of digital signal processing algorithms and machine learning techniques to enable the design of future high-capacity, intelligent optical networks, mitigating nonlinearities, reducing complexity and latency for all network operations.
- Current Projects
TRANSNET is an EPSRC-funded multidisciplinary research programme aiming to transform optical networks. Commencing in August 2018 and led by UCL, in collaboration with Aston and Cambridge Universities, TRANSNET's aim is to create an adaptive intelligent optical network - able to dynamically provide capacity where and when it is needed - the backbone of the next-generation digital infrastructure.
We are currently working on a comprehensive TRANSNET website - updates will be posted here in due course.
The COIN Project - Coding for Optical communications In the Nonlinear regime
The COIN project is a European Marie Skłodowska-Curie Innovative Training Network (ITN).
It is a collaboration project between University College London (UCL) in the U.K., Chalmers University of Technology in Sweden, Nokia (formerly Alcatel-Lucent) Bell Labs in Germany, and University of Toronto in Canada. This research project aims to develop nonlinear modulation, coding, and detection methods, tailored to the nonlinear channel, to dramatically improve the data throughput of future communication networks.
INSIGHT: Introducing Insight into the Abstraction of Optical Network Infrastructures
The INSIGHT Project is funded by the EPSRC and is a cross collaboration between UCL, University of Bristol and University of Cambridge.
The unprecedented growth of optical fibre infrastructure in recent decades has underpinned telecommunications and the Internet, making possible broadband communications, e-commerce, video-on-demand and streaming media, tele-presence and high performance distributed computing. It has dramatically changed the whole landscape of public, business and government activities, stimulating relentless traffic growth.
The INSIGHT project provides a better and more coherent strategy to sustain the growth in information-carrying digital communications infrastructure and addresses the first of the main cross cutting challenges of TI3.
Awardee: Domaniç Lavery
Awarding body: Royal Academy of Engineering
Award type: Research Fellowship
Award Amount: 0.5m GBP
Fellowship Duration: December 2016-December 2021 (5 years)
Fellowship Title: "Simplified Transceiver Architectures for High Capacity Optical Networks"
Awardee: Lidia Galdino
Awarding body: Royal Academy of Engineering
Award type: Research Fellowship
Award Amount: 0.5m GBP
Fellowship Duration: September 2018-September 2023 (5 years)
Fellowship Title: "Capacity-approaching, Ultra-Wideband Nonlinear optical Fibre Transmission System"
- Former Projects
UNLOC - UNLocking the Capacity of Optical Communications
UNLOC focused on developing new theoretical and experimental approaches to increase the capacity of optical fibre transmission technologies, which supported and transformed our economy, science, education, healthcare, transport and almost every other aspect of modern life. The programme generated path-breaking research and huge progress in the field of optical communications.
This was an extremely successful programme that made huge gains in the field of optical communications and set the tone extremely high for further research in this field.
UNLOC generated a wonderful array of exceptional published papers, conference presentations and engagements, with excellent research opportunities generated with fellow academics and key industry partners. The UNLOC website continues to be live, for a comprehensive overview of the project and outcomes, visit www.unloc.net.
COMIMO - Exploiting the bandwidth potential of multimode optical fibres
This project aimed to develop the technologies and systems required to exploit the spatial dimension of multimode optical fibre using coherent optical reception. It aims to increase the capacity of a single fibre beyond that of existing fibre communication systems in a cost effective and energy efficient manner.
Historically the optical fibre was perceived to provide “unlimited” bandwidth, however, the capacity of current communications systems based on single mode optical fibre technology is very close to the limits (within a factor of 2) imposed by the physical transmission properties of single mode fibres. The major challenge facing optical communication systems is to increase the transmission capacity in order to meet the growing demand (40% increase year-on-year) whilst reducing the cost and energy consumption per bit transmitted.
The project was funded by the EPSRC, from 25 June 2012 to 24 December 2016.
Dr Benn Thomsen, UCL - overall project lead. Leads on transmitter and receiver development, receiver based DSP and systems demonstration
Dr Tim Wilkinson, University of Cambridge - leads on efficient spatial multiplexing using Spatial light modulators
Dr Frank Payne, University of Oxford - leads on multimode fibre modelling, design and characterisation
Prof David Richardson, University of Southampton - leads on multi fibre fabrication and development of multimode optical amplification
• Gennum UK Ltd
• Oclaro Technology UK
• Rsoft Design Group
Developed and demonstrated a new ring core fibre (RCF) technology, that supports 10 spatial modes, increasing the data transmission capacity by a factor of 10 over that of conventional single mode fibre technology. The RCF developed in this project was designed to support 10 spatial channels in approximately the same size core as a conventional single mode fibre, thus dramatically increasing the capacity per unit area. In addition the transmission properties of the RCF were designed so that they reduce the required complexity of the receiver digital signal processing. During the course of this research we developed the RCF fibre, and the spatial multiplexing, optical amplification and digital signal processing components and technologies in order to demonstrate optical data transmission over ring core fibre.
Ring core fibre: A low loss 25km RCF has been designed and fabricated with a measured loss of 0.3dB/km, comparable to the 0.2dB/km loss of conventional single mode fibre. This fibre supports the transmission of 10 spatial channels simultaneouly. The ring structure of the fibre was designed to achieve a large effective index difference between mode groups, which minimises crosstalk between mode groups during propagation and simplifies the required receiver signal processing.
Spatial MUX and DEMUX: A spatial multiplexer (MUX) and demultiplexer (DEMUX) are required to efficiently couple signals from multiple transmitters into and out of the RCF. In this project we have developed and demonstrated both a flexible SLM based MUX/DEMUX and a compact and practical all fibre based Photonic Lantern MUX/DEMUX. The SLM based system is very flexible in that it lets you couple into arbitrary mode profiles and can support the launch of multiple modes, in this work we demonstrated the simultaneous launch of 6 modes. This makes it ideal for characterisation of the RCF and system performance, however the insertion loss and physical size of this system make it impractical for a deployed optical communications system. In collaboration with researchers from The University of Central Florida we developed and tested an all fibre photonic lantern that is cable of simultaneously multiplexing and demultiplexing 10 spatial modes with an insertion loss of less than 4dB and a mode selectivity better than 4.5dB.
Optical Amplification: Optical amplification is required to overcome the loss of the transmission fibre to support long distance transmission. An Erbium doped ring core fibre compatible with the passive RCF used for transmission has been designed and fabricated. A RCF optical amplifier that provides a gain of 10dB with a mode dependent gain variation of less than 1dB has been realised, however, further work is required to reduce the intrinsic loss of this fibre to increase the gain. One of the key advantages of the RCF multimode fibre over other competing multimode fibre designs for spatial multiplexing is the low mode gain variation that can be achieved with the ring core structure when using a simple pumping scheme.
Digital Signal Processing: Spatial multiplexing systems that use mode multiplexing rely on Multiple Input Multiple Output (MIMO) digital signal processing (DSP) to undo the crosstalk between the different spatial channels that occurs due the imperfections in the optical components. Typically the required signal processing scales with the square of the number of spatial channels and linearly with transmission distance. As such approaches to reduce this square law scaling and distance scaling in complexity are needed to make these systems practical.
The ONG optical networks and systems laboratory is the leading experimental laboratory in the world. It has unique facilities for the state-of-the-art optical communications systems, networks and device research and test & measurement equipment.
Supporting experimental work is a powerful capability for analytical and numerical modelling of signal propagation in fibre to predict performance and refine experimental measurements. The ONG lab enables this by:
Transmission and network test-bed
The transmission & network test-bed is based around an optical fibre recirculating loop and has comprehensive signal generation and detection capabilities. Current version of digital transmitter includes DACs with sampling rates up to 92GSa/s, capable of generating advanced modulation formats with symbol rates from 6 GBd to 92 GBd. Both direct- and coherent detection is used with receiver bandwidth of up to 67 GHz and a range of sophisticated digital signal processing algorithms to compensate for and mitigation of both linear and nonlinear impairments.
Current capabilities include the generation of polarisation division multiplexed, 64QAM, 256QAM and 1024QAM with probabilistic and geometric shaping of up to 1 Tbit/s across 100 nm of optical bandwidths (approx 10 THz). To date, experiments over these bandwidths, applying nonlinearity compensation techniques at the transmitter and receiver have helped to achieve record-breaking throughputs.
Research into advanced optical transceivers has included a range of direct-detection techniques, including the Kramers-Kronig receiver, directly-modulated lasers, and digital pre-compensations, in conjunction with subcarrier modulation and low complexity PAM signals. The current of research focuses on intelligent transceivers, enabled by new transceiver design and machine learning algorithms that help transform future optical networks for the cloud.
State-of-the-art fibre processing and splicing facilities
The ONG laboratory extends to the CONNECT laboratory and is a node in the UK National Dark Fibre Infrastructure Service (NDFIS). The CONNECT lab is a home to the ONG research on data centre interconnects and novel optical switching. Research involves data disaggregation, real time optical switching and clock and data recovery on nanosecond time scales.
The ONG laboratory is equipped with state-of-the-art fibre processing and splicing facilities that supports a wide range of research and collaborations related to special optical fibres. The optical fibre laboratory has the capability of cleaving, tapering, splicing and recoating optical fibres with diameter from 80 µm to 1.2 mm, which allows ONG to research on special fibre (e.g. polarization maintaining, multimode, multi-core fibre, hollow-core) based optical systems.
In collaboration with academic and industrial partners, we have returned to the investigation of optical devices including high-speed semiconductor lasers, integrated silicon transceivers, and optical signal processing methods. A new device test-bed allows for butt and vertical coupling optical signals into integrated optical chips, supports system verification and innovation in integrated optical devices.
ONG has strived to ensure that our research should provide leadership through applications. It has successfully underpinned many now-deployed commercial systems, leading to dramatic advances in system capacity.
We are extremely proud to have collaborated with over 50 leading industrial and academic laboratories who have actively supported our work in a variety of ways. Throughout the life of ONG, our research has underpinned the evolution of optical communications and networks. Some of our current collaborations include Microsoft on Optics for the Cloud, Sumitomo on next-generation multi-core fibres, NICT (Japan) on future optical networks, Mitsubishi Electric Research Laboratories (MERL, USA) on advanced modulation and coding, and numerous others, see for example the TRANSNET programme.
Join our team
We have a number of fully-funded PhD studentships starting in 2019/2020, in collaboration with our industry partners.
We are looking for academically excellent and motivated students, with Bachelor’s or Master’s degree in Physics, Electronic & Electrical Engineering or Computer Science. You should have an interest in optical communications and/or networks.
Although it is not required, hands-on lab experience in optical communication research is an advantage, as are excellent computer and lab skills. Other useful skills/knowledge include optical fibre communications, digital signal processing, optics and nonlinear optics, mathematical modelling and machine learning. Some of the current projects address challenges in:
- Optical networks for the ‘cloud’
- Low-complexity, intelligent transceivers – simplifying digital signal processing
- Design of ultra-wideband, high-speed optical fibre transmission systems & networks
- Novel optical amplifiers for future optical network applications
- Optical signal processing for communications and instrumentation
Funded positions are for UK/EU students only. Applications from *exceptional* overseas students may be considered. If you are interested, we’re accepting applications all year around – please send your CV and names of 2 referees to Prof Polina Bayvel (email@example.com), or apply directly (read research guidance in full, link to apply at the bottom of the page), marking your application TRANSNET/Polina Bayvel.
Fully–funded PhD Studentship – Intelligent Optical Networks enabled by Machine Learning
PhD Studentship in collaboration with HUBER+SUHNER Polatis: Towards 1000-times capacity increase for Self-x Cloud network infrastructure