The University of Kent outlines how it is leading research focused on the radio access network of future mobile communications…
The next, 5th generation of mobile communication networks is expected to support ultra-high-bit-rate services, such as for high-definition video, low-latency, for gaming, self-driving vehicles, tele-surgery, etc., and the vast interconnection of devices which will be the internet of everything. Substantial changes will be needed in the way mobile networks operate to support these diverse demands, with the technology advances being made in current research around the globe, including that funded by the European Union’s Horizon 2020 (H2020) programme, vital for success.
While many developments have been occurring for the high-speed core of the networks, for data centres and for increased “softwarisation” to enhance the organisation and optimisation of networks, it is in the radio access, which must connect the many, and often bandwidth-hungry users, that revolutionary changes may be needed. It may seem odd that a Professor of Optical Fibre Communications, such as Nathan Gomes at the University of Kent should be leading projects contributing to such advances, but fibre optics has a key role to play in future radio access networks. “5G will need numerous radio access nodes, closer to the user to provide high bandwidths – only fibre optics can bring the necessary bandwidth to these nodes, for many of the services envisaged”, says Professor Gomes.
In projects such as iCIRRUS (funded by the EU’s H2020 programme) and NIRVANA (funded by the UK’s Engineering and Physical Sciences Research Council), the integration of the optical fibre distribution within the radio access network is being investigated. The fibre distribution and the radio access are seen and designed as a whole. The projects use the concept of a virtualised radio access network (or vRAN), where the communication functions that need to be carried out can be distributed as required in different physical entities, dependent on current demands and in a manner that optimises user experience and overall network performance. In iCIRRUS and NIRVANA, an Ethernet-based fibre optic distribution network (a “fronthaul”) is used to interconnect these flexible physical entities, Ethernet offering a high potential for shared infrastructures and low cost.
The vRAN is a new concept going beyond the centralised, cloud-RAN, which has quite recently been of significant interest to mobile operators. Whereas the cloud-RAN envisages centralised operation, the vRAN permits functions to be performed where necessary. This can bring benefits to device-to-device (D2D) communications, where users are allocated network resources to communicate directly, unburdening the rest of the network. Placing some control locally in a vRAN could enhance the setup processes necessary. Also, users who are between radio access points can beneficially send/receive signals to both (it could be more than 2) while those close to a radio access point do not need to. This affects where the signals for such users should be processed – a flexibility enabled with a vRAN. In iCIRRUS, an 11 partner project, a University of Kent D2D team collaborates with industry partners such as InterDigital in the UK and Primetel, Cyprus, while an Ethernet fronthaul team collaborates with partners such as ADVA Optical Networking, Germany and Orange, France.
Over the last couple of years, standardisation work has started for using Ethernet in the RAN fronthaul. However, up to now, it has not taken into account the vRAN and its flexible allocation of functions to physical entities. “Among EU projects, iCIRRUS is unique in its proposal and detailed investigation of the use of Ethernet within the new vRAN”, says Professor Gomes. “The iCIRRUS partners have a particular opportunity to contribute to new, flourishing international research and standardisation efforts. It is a project like no other I have been involved in, with rapid developments to keep track of from industry and academia across the globe. Conversely, it also means that there is a great deal of international interest in what we are doing; iCIRRUS partners are disseminating the progress being made in the project in meetings of many international standards bodies and industry groups, as well as at major conferences”.
What are the next steps? They mainly focus on validation of the key concepts. In fact, there are considerable challenges in making an Ethernet fronthaul work, as the packetisation and variable buffering delays that typically occur in these networks could be catastrophic within the vRAN. After laboratory experiments, for example, at University of Kent, Faunhofer-HHI (Berlin), and ADVA (Munich) it is intended to use the operators in the consortium, such as Orange and Telekom Slovenije, to validate the technology in their test networks. Watch this space, as some of the validation setups should be opened for a public viewing at the end of the project (late next year).
Fronthaul:
Fronthaul is used to describe the links between base stations and radio units in a mobile network (as opposed to backhaul which connects the base stations to the core network). Currently, fronthaul technology is mainly based on an industry specification called CPRI, which transports digital samples of radio waveforms, and leads to bit-rates orders of magnitude higher than the user data bit-rates. Projects such as iCIRRUS and NIRVANA propose transporting different digital signals with reduced bit-rate requirements. Other projects, such as the EU-Japan project RAPID propose analogue signal transport to reduce bandwidth requirements. For more information, visit the project websites.
iCIRRUS: http://www.icirrus-5gnet.eu
NIRVANA: http://www.intelligent-nirvana.net
RAPID: http://www.rapid-5g.eu
Professor Nathan Gomes
Professor of Optical Fibre Communications
University of Kent
+44 1227 823719
N.J. Gomes@kent.ac.uk
Please note: this is a commercial profile
