The PREDICT-6G consortium consists of 17 partners from seven countries working together to develop a multi-technology, multi-domain data plane that revises the reliability and time-sensitive design features existing in current wired and wireless standards.
While we all work towards a common goal, each partner has a specific role with concrete milestones. Thus, we have asked them about the highlights of PREDICT´s first year and what they expect to achieve in 2024.
UNIVERSIDAD CARLOS III DE MADRID
At UC3M, we have been mostly focused on developing several components of the MDP, specially on developing an open source XDP based bridge, which can be used to develop novel scheduling policies. We have also deployed multiple technologies to start building the MDP, including Wireless TSN, Ethernet-based TSN, Open Air interface 5G network, and programmable P4 switches.
During the second year of the project, through this infrastructure, we will focus on developing the WP3 components and prove the new approaches for the PREDICT-6G data plane.
In the first year of PREDICT-6G, ATOS+ATOSP’s main focus was the design of the AI-driven Inter-domain Control-Plane (AICP). In particular, ATOS+ATOSP led the efforts related to the architecture design of the inter-domain and network control mechanisms, assuming the responsibility of defining Service Automation Management Services. Additionally, the company was an active contributor to the composition of the AI/ML layout, taking the lead in the development of the AI/ML framework, which will be used by the project as the basis for building complex AI/ML algorithms.
Looking towards PREDICT-6G second year, ATOS+ATOSP will concentrate their efforts in the implementation of the Service Automation and the AI/ML framework modules, which are considered core services for showcasing the potential of the AICP in PREDICT-6G Proof-of-Concepts (PoCs).
This year was very important for partners to build a close relationship that would underpin all the project work, including the implementation of communication and dissemination strategy. Consolidating the PREDICT-6G branding, setting up the digital tools – website and social media – and overall, positioning the project in the SNS JU, were some of the main accomplishments in the last 12 months. Moreover, AUSTRALO established the first contacts with relevant stakeholders in the SNS community with very positive results.
In Year 2, AUSTRALO will continue to foster the PREDICT-6G community. In addition to the communication activities, the engagement with relevant stakeholders through targeted actions will accelerate to ensure the impact of the project. The dissemination of outcomes, as PREDICT-6G matures and the first results emerge, will be one of the priorities. We expect to support PREDICT-6G to become a reference in the SNS system architecture.
During the first year of PREDICT-6G, our main objective and area of work has been to identify security threats to predictive, multi-domain and technological 6G mobile networks, and corresponding challenges that need to be assessed through specific mechanisms and protocols. Following the completion of this work, next year we will focus on designing and developing mechanisms and protocols to provide the necessary security solutions.
CONSIGLIO NAZIONALE DELLE RICERCHE
In this first year, the CNR activities started with a comprehensive exploration of the State of the Art (SoA) in deterministic networks, with specific focus on Time-Sensitive Networking (TSN). The primary goal was to identify an effective methodology for assessing the performance of this emerging technology, particularly in terms of latency and jitter. Collaborative efforts were undertaken with all project partners to establish the architecture and define relevant case studies. In this sense, the definition of automation and robotics scenarios in line with the PREDICT-6G industrial case studies was done so as to allow gaining insight into the TSN application to information spreading and edge computing.
Subsequently, CNR people conceptualised a sophisticated software architecture designed to evaluate latencies in both individual devices and network infrastructures, incorporating scenarios with and without running synchronisation systems. At the same time, rigorous prototype implementation and test activities were conducted to assess TSN performance over WiFi. To this regard, some preliminary results of such activities have been presented in a paper published in the Proceedings of the IEEE ETFA 2023 International Conference.
In 2024, the CNR will investigate the framework described above to assess the performance of devices and analyse the results to create virtual counterparts for use in the Digital Twin. Installing a prototype hybrid wired/wireless/5G network to conduct test campaigns on the components of the PREDICT-6G architecture.
To perform realistic simulations of sensitive scenarios to study useful scaling laws and suggest control strategies to improve performance in deterministic communication and task completion.
In the first year of the project, the team dedicated their efforts to several key initiatives. A primary focus was the in-depth study and analysis of the Smart Factory use case. Internal workshops played a crucial role in establishing a first draft of the architecture framework needed for said use case. The project achieved a significant milestone with the delivery of D1.1, titled “Analysis of use case and system requirements.” Additionally, the team delineated the requirements for both the Multi Data Plane (MDP) and the Control Plane.
As the project advances into its second year, the central emphasis shifts towards the design of the use case integration. This will be accompanied by the inaugural testing of the use case within the 5TONIC Open Lab. The outcomes of these tests will be documented to facilitate further experimentation and analysis in subsequent phases.
During the first year, the Consortium helped to define the architecture and interfaces between the AI-based control plane and the data plane innovations (e.g., exploitation of new Wi-Fi features for improved determinism). This outcome was a product of a dialog between experts in determinism in the different wired and wireless domains, addressing the challenges to integrate the available options in 3GPP, IEEE and IETF standards.
The implementation and integration of technologies in the testbed such as Wi-Fi based TSN, as well as E2E and domain-specific management services such as AI-based slicing will be central in 2024. The evaluation of the target KPIs defined in the project can then be assessed.
Working with so many talented individuals and impactful organisations towards the definition of deterministic communications in 6G has been the highlight of the first year.
In Year 2, InterDigital will prove the novel concept of Data Unit Groups,. This extension to IP headers will allow to treat a set of packets carrying a single Application Data Unit collectively, e.g. to map the information inside the IP header extensions to PDU Set Markers in 3GPP networks or to place them into specific TSN switch queues and/or to pre-empt them if deemed necessary to guarantee their QoS constraints. This innovation will be demonstrated an Integrated Sensing and Communication use case. 5TONIC serves as the TRL4 environment to validate the proof-of-concept.
The first year started in-depth analysis of the technical requirements imposed by the three use cases that will be demonstrated at the end of the project and the subsequent design of the system architecture. This was a huge work that was split in two parts: one focusing on the data-plane, and another on the control/management-plane. This design resulted in a PREDICT-6G architecture that is modular, with clear separation between technology domain management and end-to-end service management, which in turn enables extendibility to any new network technology without changing end-to-end principles.
The realisation of all PREDICT-6G components and their integration into a coherent system will be the priority in 2024. The implementation work has already started on a small scale, with a few code drops delivered at the end of 2023.
As far as POLITO activity was concerned, the first year of the project has been marked by acquiring competence in creating seamless deterministic services across network domains, employing a blend of telecommunications standards and innovative Wi-Fi technologies, all unified under a flexible OpenAPI-driven control architecture.
The main focus of the second year will be on providing deterministic guarantees in WiFi networks by exploring the potential of a comprehensive 802.11 scheduler that incorporates time, frequency, and space dimensions, with an emphasis on dynamic adaptation to variable traffic and radio channels. The objective will be to efficiently allocate deterministic traffic while maximising throughput for other traffic, incorporating frequent scheduling updates and considering the integration of preamble puncturing for future-proofing against emerging WiFi standards.
UNIVERSITAT POLITÈCNICA DE CATALUNYA
In its inaugural year, the UPC group achieved significant milestones in different key areas: dynamic scheduling for radio resource management in mixed traffic types for private 5G networks, and AICP Management Services (MSs) to support E2E services provisioning and maintenance.
Dynamic Scheduling for Radio Resource Management in Mixed Traffic Types for Private 5G Networks involved substantial progress in the domain of dynamic scheduling and OFDMA radio resource management in the context of mixed traffic types (Isochronous TSN, Asynchronous TSN and Best Effort (BE)) within private 5G networks deployed in industrial settings. This entails the effective allocation of resources and admission control in response to the varying communication demands of diverse traffic types. The outcomes of this effort include improved spectral efficiency, reduced latency/jitter and enhanced reliability, all crucial factors for the successful integration of 5G technologies in private networks and the support of a range of applications in the industrial domain.
On the other hand, UPC actively contributed to the design and preliminary implementation of AICP MSs, such as Digital Twin for KPI evaluation and Path Computation, both at technology domain level and at the E2E level. Both MSs are essential for the deterministic services provisioning assuring their requirements and represent significant components of the PREDICT-6G AICP.
In conclusion, the first year has been marked by commendable progress. These achievements reflect the initiative’s commitment to advancing research in the field of deterministic services, laying a solid foundation for continued exploration and innovation in the months to come.
Building on the achievements of the first year, the second year of PREDICT-6G considers the following focused areas to further advance the research and development:
- Advanced Dynamic Scheduling and Resource Management in Mobility Scenarios: This involves refining and enhancing dynamic scheduling algorithms for radio resource management, placing a particular emphasis on the allocation of resources in mobility scenarios where handovers should not affect the continuity of the service. To that end, we will consider the resource allocation for the Dual Active Stack Protocol handover. We will incorporate machine learning techniques to enable the network to adapt to UL/DL traffic variability and learn from dynamic industrial environments, ensuring improved performance.
- Development and integration of a Digital Twin module for KPI evaluation as part of the AICP.
- Development and integration of a Path Computation modules as part of the AICP.
- Integration on the Advanced Dynamic Scheduling and Resource Management in the Digital Twin: Explore the integration of the studied technologies to augment the capabilities through the integration of the schedulers in the E2E digital twin tool. Investigate how these technologies can synergize with dynamic scheduling approaches to enhance reliability, latency and real-time decision-making about the admission and rejection of traffic flows.
- Standardisation and Industry Adoption: Engage with pre-standardization bodies and industry stakeholders to contribute insights and findings to the development of 6G standards.
If you want to stay updated about PREDICT-6Gt, subscribe to our newsletter and follow us on Twitter and LinkedIn!