5G PICTURE

Link to project website: http://www.5g-picture-project.eu/

Social Media: Twitter @5G_PICTURE

Contact: 5g-picture-Contact@5g-ppp.eu

Objective

5G-PICTURE will design and develop an integrated, scalable and open 5G infrastructure with the aim to support operational and end-user services for both ICT and “vertical” industries. This infrastructure will rely on a converged fronthaul and backhaul solution, integrating advanced wireless access and novel optical network domains. To address the limitations of current solutions, 5G-PICTURE will adopt the novel concept of Disaggregated-Radio Access Networks (DA-RANs), allowing any service to flexibly mix-and-match and use compute, storage and network resources through HW programmability. It will also rely on network softwarisation to enable an open reference platform instantiating a variety of network functions and adopt slicing and service chaining to facilitate optimised multi-tenancy operation.

Achievements

The 5G-PICTURE results are being validated in three demonstration activities:

Railway use case (November 2019, Barcelona, Spain)

The 5G-PICTURE Railway demonstration is Europe’s first 5G Rail deployment in an operational environment. The demo installation at Ferrocarrils de la Generalitat de Catalunya (FGC) comprises approximately 1.5 km of track providing 1 Gbit/s throughput to a train running up to 90 km/h. The demo solution designed, developed and integrated during the project execution is based on a mmWave RAN providing vehicle to infrastructure connectivity and a wireless access solution inside train. This model implements the 5G-PICTURE architecture for railways, for a cost efficient and multitenant 5G implementation, to facilitate 5G deployment in this vertical.

The developed solution is based on (1) Blu Wireless Technology self-directed mmWave beams between train and track poles, to provide multiple wireless train to track connections, (2) state of the art ADVA Optical Networking ITU-T G.698.4 agnostic passive optical connectivity between poles, integrated into project purpose designed outdoor cabinets from COMSA Industrial, a HW programmable mobility server solution from Consorzio Nazionale Interuniversitario per le Telecomunicazioni (CNIT) to preserve session continuity, (4) a 100GE aggregation/protection layer to provide connectivity to the railways core network, and (5) a TCN network design from COMSA.

The solution is designed to support multiple 1 Gbit/s throughput, end to end latency below 2 ms and the support of precise time synchronization with lowest possible cost. This solution is a clear contribution to the FRMCS vision for the implementation of a 3GPP based solution for all railway’s telecom needs.

In terms of services 5G-PICTURE railways trial has demonstrated services from the three defined types by FRMCS: (1) forward-looking low latency critical video over a 5G railway experimental testbed, (2) HD video streaming with the high-throughput traffic originated by the Business services inside the train, (3) Emulated Performance services with different traffic patterns (bandwidth, latency, packet jitter, etc.) that can be access via internal Wi-Fi. A large number of tests were completed for both scenarios: stationary train and moving train. The demo shows E2E performance up to 2 Gbit/s (DL+UL) to the moving train with average latencies of 2.35 ms and power consumption per km of 200 W.

Smart City demonstration (March 2020, Bristol, UK)

The University of Bristol test network has been used for the 5G-PICTURE smart city use case demonstration. The main objective of this demonstration has been to showcase the multitenancy capability of 5G-PICTURE disaggregated transport layer and, at the same time, its capability to meet the forthcoming 5G services’ performance requirements. For this purpose, 5G-PICTURE equipment and its disaggregated transport solution was integrated within the 5GUK test network spread across three different physical locations, comprising (among other infrastructure) 5G NR, LTE-A and Wi-Fi nodes at the access network part, a 5G NSA/LTE core as well as MEC and cloud compute resources. IHP GmbH integrated three devices operating in the unlicensed 60 GHz band to the test network, serving as a backhaul mesh segment. Xilinx Dresden brought an active massive MIMO Antenna Proof-of-Concept platform Radio Unit (RU), along with a Central Unit (CU) server solution emulating the base band processing along with a spectrum analyser as receiver emulating a UE function, serving as fronthaul transport stream. All these backhaul and fronthaul transport streams have been integrated with the Time-Shared Optical Network (TSON) technology developed by the University of Bristol in the framework of 5G-PICTURE. This technology was employed to enable mapping of multitenant traffic across infrastructure domains.

The 5G-PICTURE Smart City demonstration successfully showcased the capabilities of the aforementioned deployment using two use cases: smart city safety and virtual reality. The critical performance KPIs associated with these use cases are radio network node capacity, end-user throughout and latency; therefore the demo focused on the evaluation of these KPIs. The evaluation results reveal that the 5G-PICTURE solution successfully met the expected KPIs targets for these use cases and for 5G services in general. In addition to the use case, a mMIMO RU demonstration successfully validated the benefits both of Sub-6 GHz mMIMO, as well as the functional split 7.2 and corresponding fronthaul interface.

Stadium demonstration (March 2020, Bristol, UK)

This use case for the 5G-PICTURE Stadium demonstration at a stadium in Bristol UK was presented. The purpose of 5G-PICTURE was to augment an existing production network in a stadium in Bristol, UK, to a 5G network and an implementation of the 5G Operating System (5G OS) architecture developed in 5G-PICTURE in a complex production network environment.

The demonstration focused on addressing the following 5G themes for the mega-event/stadium vertical:

• Application aware, programmable network over heterogeneous hardware.
• Differentiated treatment of the application traffic using slices.
• Service resilience using slices in a multi-connectivity link scenario for Wi-Fi and high capacity wireless access technologies, such as Massive MIMO.

For the use case a real-time, crowd-sourced video production application called Watchity was used. Watchity comprises an application for capturing and uploading video/audio from an end user device and a cloud hosted video production platform, which in the case of 5G-PICTURE, ran on Amazon Web Services. The demonstration successfully showed how the 5G-PICTURE intelligent network autonomously detected and prioritised video traffic in order to maintain QoE.

The demonstrator also proved the deployment capability to:

• Achieve 20-25 Mbit/s per user.
• Slice provisioning time below 40 seconds (37.8 seconds).
• Slice instantiation and reconfiguration (HO) in the order of some seconds – less than 40 seconds (compared to hours required currently, as they require manual interaction).
• RTT of 40 ms even during the slice reconfiguration (HO) phase.
• Achieve low network service deployment and termination times – below 5-6 seconds for services consisting of 4 VNFs.
• Service Termination time below 5-6 seconds.
• Service Deployment time in Compute domain below 5-6 seconds

For more details on the 5G-PICTURE trials, please see 5G-PICTURE deliverable D6.3 Final Demo and Testbed experimentation results.

Project Participants: