CORDIS - EU research results

intelligent Converged network consolIdating Radio and optical access aRound USer equipment.

Periodic Reporting for period 3 - iCIRRUS (intelligent Converged network consolIdating Radio and optical access aRound USer equipment.)

Reporting period: 2017-01-01 to 2017-12-31

iCIRRUS proposes an intelligent Ethernet-based fronthaul for the Radio Access Network (RAN) of 5th Generation (5G) and beyond mobile communications. The fronthaul is a core component of the cloud- or virtualized-RAN, C-RAN or vRAN, that is now seen as essential for such future mobile systems. The intelligence, enabled by the use of Ethernet and probing the resulting fronthaul network, can also enhance the operation of other, user-facing applications. In iCIRRUS, cellular-assisted device-to-device (D2D) communication is targeted, as some network functions are placed closer to users and could aid localisation, for example, as is mobile cloud computation, which again may benefit from placing functionality closer to users in the C-RAN/vRAN. Further, an aim is to examine how the intelligence gathered can be used to optimise network performance, for the fronthaul and its effect on radio performance, and for the D2D and mobile cloud operations.

The iCIRRUS project has delivered a number of testbed demonstrations that have separately verified key performance targets identified for the network architecture. It has also delivered a final, integrated showcase demonstration that combines five different fronthaul/midhaul/backhaul streams through an Ethernet time-sensitive networking aggregator with support for timing and synchronisation, together with mobile cloud/clone use cases for video transcoding/streaming and task offloading services.

Overall, the Ethernet fronthaul concept proposed in the project has been validated through the measurements that verify key performance indicators can be met, and through analysis and simulation. This evolved fronthaul, supporting a virtualised RAN will be a necessary component if future 5G deployments. By focussing on Ethernet and addressing its main challenges for the fronthaul, iCIRRUS has made a significant contribution to research and innovation in this area.

The mobile cloud/clone operations have been demonstrated in the context of a Unified Communications use case, which has strong business potential. Significant theoretical analyses have been published in the highest-raking IEEE journals clearly exposing the energy savings inherent in the mobile cloud/clone, C-RAN/vRAN and D2D proposals.

The work has influenced standards bodies through direct presentations or through indirect interactions, and is clearly aligned with the current thinking in important bodies such as 3GPP.

The work has met all of the objectives set out at the beginning of the project:

1. Energy efficiency enhancements through the C-RAN/fronthaul, mobile cloud and D2D, together with joint resource management.
2. Feasibility of the Ethernet fronthaul and convergence with fixed-access (Passive Optical Networks - PONs).
3. D2D spectral and energy efficiency reduction through use of the evolved RAN and a new signalling scheme for device discovery.
4. Spectrum and energy efficiency for the mobile cloud, particularly through task and communication offloading.
The initial work in the iCIRRUS project focussed on obtaining a consensus for the overall system architecture. The project examined scenarios (test cases) from the METIS project and other 5G white paper proposed scenarios, but the major consideration came from the realisation that current fronthaul technology could not be simply extended to meet the requirements of 5G. Instead, a radically new approach was required. This realisation was being made independently by others. The main effort in our overall systems work package over the first 6 months of the project, was to consolidate this thinking, defining clearly that a new functional split in the fronthaul was necessary and that the split could be implemented at different points. The effect of this new split on the overall mobile system, including the D2D and mobile cloud, was considered and the challenges that it brought were identified.

The next phase of the work was the more detailed examination of these challenges. For the fronthaul, the different split point options were compared and the requirements of each, in terms of data rates and latency/delay variation requirements were outlined. Similarly, the different functional splits offered opportunities and challenges to the D2D and mobile cloud operation, and these were more closely studied. Specifications for their possible operation were made. Overall, a new flexible vRAN/fronthaul architecture with possibilities for real convergence with other types of network was formulated – this is shown in Fig.1.

The work carried out resulted in several publications, including an invited paper on the initial Ethernet fronthaul concept, in the second year of the project, and as a result of the more detailed studies during this time, 10 high-quality journal papers were published in the third year of the project, covering Ethernet fronthaul and PON convergence, C-RAN energy efficiency and mobile cloud energy efficiency studies. The final year's work has led to the overall conclusions stated in the previous section.

The exploitation of results is demonstrated by 5 patent filings, standards bodies contributions and detailed business planning from all industry partners using the lean canvas template.
The project has made the most detailed examination of fronthaul requirements for 5G, particularly considering new functional splits and comparing their requirements. For the D2D work, a novel proposal for grouping devices that benefits from the functionally split C-RAN architecture has been made. Proposals for balancing and optimising between operations in a central cloud and those in a temporary storage in edge clouds supported in a C-RAN type network have also been made.

The project has clearly made steps beyond the state-of-the-art, evidenced by its top journal paper publications and invitations to international conferences in the areas of fronthaul, particularly on Ethernet and meeting the challenges of its use, on fronthaul Ethernet-PON convergence, on D2D communications, mm-wave technology for mesh networking, and mobile cloud/clone communications.

The project has measured itself using KPIs, including business/societal KPIs that have been defined for EU 5G projects. Our showcase demonstrator in TS laboratories, and an earlier demonstrator in the laboratories of Orange, contribute to the KPI requesting more proof of concept demonstrations. To an extent this is also true of the smaller individual demonstrations, contributions to standards and industry events. Reduction in energy consumption per service is also a societal KPI that is supported by the whole iCIRRUS concept. A final societal KPI that has been supported by the iCIRRUS work is the stimulation of new economically viable services: this has been met on a number of fronts, as detailed in the exploitation plans of the industry partners. In terms of public, user services, the plans of Wellness Telecom for new service offerings based on the research in iCIRRUS are particularly pertinent.

The project has used its website and social media to communicate with more general audiences, including the public, and Telekom Slovenije has recently made a YouTube video which presents a high-level overview of the project's achievements in the context of the increasing interest in 5G rollouts.
Figure 1
iCIRRUS logo