PREDICT-6G

A look back at the first six months of PREDICT-6G

By Péter Szilágyi, PREDICT-6G Technical Manager

It’s been six months since PREDICT-6G kicked off, and the wheels – prepared carefully during the planning – were set in motion. Within that time, we have built a well functioning team, gained remarkable momentum, and started to fill our pipeline of technology and innovation. This article reflects on the key achievements so far, the status of the project, and the next steps – from a technical perspective.

What we have achieved

The first period of the project focused on setting the landscape, collecting and structuring ideas, discussing and documenting our approach to multi-domain deterministic networks and services, and roadmapping the underlying technical work. We have established solid understanding of the use cases (such as smart manufacturing and industrial critical communication) requiring deterministic services across multiple domains. For each use case, we studied the end-to-end communication requirements, the capabilities of the devices, the types and roles of the actors and their anticipated demand and traffic mix. The findings were consolidated and generalized to produce system level and service level KPIs and requirements, serving as design principles and capabilities towards the PREDICT-6G system. Based on the requirements, we have already created the first architecture blueprint of the project, including both the multi-technology multi-domain data plane and the AI-driven inter-domain control plane, as well as the means of their interworking. On the groundwork side, we have established a roadmap for the two Open Labs where the PREDICT-6G technology and innovations will be implemented and demonstrated through the selected use cases. Finally, we have created two important pieces of deliverables: our Data Management Plan (DMP), which defines the standards for preparing, publishing and maintaining open access to all types of data, including documentation, measurements and source code to be produced by PREDICT-6G; and our first version of the communication, dissemination, standardization and exploitation strategy plan.

Where we are now

We are in a busy schedule paved with deliverables on all of our research directions. We are just releasing D1.1, our first technical report summarizing the use cases, requirements and initial architecture of the PREDICT-6G. I recommend reading this document for those who would like to get familiar with the project’s technical line of thought and innovation areas, as these aspects already started to manifest in this work. In parallel to finishing D1.1 under WP1, we are cooperating very closely between two of our other technical work packages: WP2, which is defining the deterministic technologies for data planes and cross-domain data plane integration; and WP3, which is defining the automation framework on the control and management plane to self-orchestrate and autonomously assure end-to-end deterministic services. These two WPs are expected to be the busiest ones in the next four months, as they work together to co-create two sets of dependent technologies. On the one hand, in WP2, to expose programmable data plane capabilities from within specific network technologies such as 3GPP, IETF DetNet/RAW, and Wi-Fi; and on the other hand, in WP3, to utilize those exposed capabilities to autonomously fulfil and assure the end-to-end services. Additionally, WP4 takes off in July, to start working on system integration aspects that would bring all PREDICT-6G technical components into the same autonomous framework.

Looking ahead

Summer will be hot for PREDICT-6G, and not only for the season. We will produce our next two deliverables, which are going to consolidate part of the effort we are currently putting into the cross-WP2-WP3 work. To be released at the end of August, D2.1 will descend deeper into the data plane technologies, whereas one month later, D3.1 will report the first results on the control and management plane technologies, already leveraging the capabilities of our data plane. These deliverables will maintain close coherence and context with each other and with D1.1, so that interested readers can easily navigate the breadths and depths of the PREDICT-6G technology. These two upcoming deliverables will also set up the forward path to their second versions, which are both due near the end of this year, going even deeper into their subjects. By that time, PREDICT-6G will have created and released substantial technical capital that will be a foundation for starting lab work in 2024. 

Summary

Nowadays are intense yet interesting times in PREDICT-6G. Our momentum accelerates, the frontier of the research broadens, and the number of ongoing activities increase. While this means that work is split up and task forces are focusing on specific areas of PREDICT-6G to maintain efficiency and productivity, we take special care to leverage cross-WP and cross-partner expertise as we drive along our micro-objectives. All in all, we have an excellent team working together on a research project with a great aim – and that is all that’s needed.

If you want to always stay updated about our project, subscribe to our newsletter and follow us on Twitter and LinkedIn!

by PREDICT-6G

How Networks Can Help Machine Learning to Becoming (Truly) Pervasive

By Prof. Carla Fabiana Chiasserini, Politecnico di Torino, Italy

Prof. Carla Fabiana Chiasserini, member of the PREDICT-6G Consortium on behalf of Politecnico di Torino, Italy, highlights the challenges that the ubiquitous use of machine learning is posing and how the PREDICT-6G project is developing solutions to make it sustainable.

Machine Learning (ML) is all around: it is becoming an essential component of many user applications and network services. However, we all know that training and executing a ML model may exact a significant toll from the computational and network infrastructure due to its high resource demand. Consequently, current implementations of ML operations are heavy energy consumers, which makes the pervasiveness of ML we are witnessing not sustainable.

PREDICT-6G is committed to find breakthrough approaches to take and solve the challenge. Specifically, it has tackled the use of services for the optimal configuration of virtualized radio interfaces and of user applications at the network edge, for which ML can be the problem and the solution at the same time. 

Network Function Virtualization (NFV) and edge computing are indeed disrupting the way mobile services can be offered through mobile network infrastructure. Third parties such as vertical industries and over-the-top players can now partner up with mobile operators to reach directly their customers and deliver a plethora of services with substantially reduced latency and bandwidth consumption. Video streaming, gaming, virtual reality, safety services for connected vehicles, and IoT are all services that can benefit from the combination of NFV and edge computing: when implemented through virtual machines or containers in servers co-located with base stations (or nearby), they can enjoy low latency and jitter, while storing and processing data locally. 

The combination of NFV, edge computing, and an efficient radio interface, e.g., O-RAN, is therefore a powerful means to offer mobile services with high quality of experience (QoE). However, user applications are not the only ones that can be virtualized: network services such as data radio transmission and reception are nowa- days virtualized and implemented through Virtual Network Functions (VNFs) as well; and both types of virtual services, user’s and network’s, may be highly computationally intensive. On the other hand, it is a fact that computational availability at the network edge is limited. It follows that in the edge ecosystem, user applications and network services compete for resources, hence designing automated and efficient resource orchestration mechanisms in the case of resource scarcity is critical. 

Further, looking more closely at the computational demand of virtualized user applications and at that of network service VNFs, one can notice that they certainly depend on the amount of data each service has to process, but they are also entangled. As an example, consider a user application at the edge and (de-)modulation and (de-)coding functions in a virtualized radio access network (vRAN). For downlink traffic, the application bitrate determines the amount of data to be processed by the vRAN; on the contrary, for uplink traffic, the data processed by the vRAN is the input to the application service. A negative correlation, however, may also exist: the more data compression is performed by a user application, the higher its computational demand, but the smaller the amount of data to be transmitted and the less the computing resources required by the vRAN. In a nutshell, a correlation exists between the amount of data processed/generated by virtual applications at the edge and network services VNFs, and such correlation can be positive or negative depending on the type of involved VNFs. Experimental tests performed within PREDICT-6G clearly show such correlation. 

Then, owing to the complex involved dynamics, PREDICT-6G has developped a scalable reinforcement learning-based framework for resource orchestration at the edge, which leverages a Pareto analysis for provable fair and efficient decisions. The developed framework, named VERA [1], meets the target values of latency and throughput for over 96% of the observation period and its scaling cost is 54% lower than a traditional, centralized framework based on deep-Q networks. 

[1] S. Tripathi, C. Puligheddu, S. Pramanik, A. Garcia-Saavedra and C. F. Chiasserini, "Fair and Scalable Orchestration of Network and Compute Resources for Virtual Edge Services," in IEEE Transactions on Mobile Computing, doi: 10.1109/TMC.2023.3254999.

If you want to always stay updated about our project, subscribe to our newsletter and follow us on Twitter and LinkedIn!

by PREDICT-6G