Project Content
One key innovation in the 6G era will be the new design of the system architecture to support the extreme cloud continuum, where network functions from different 6G network segments can be composited flexibly and dynamically based on service needs in diverse cloud environments.
With this in mind, 6G-Cloud aims to go far beyond the current 5G architecture design with disruptive and innovative approaches and solutions. It will research, develop, and validate key technologies to realize an artificial intelligence (AI)-native and cloud-friendly system architecture atop the cloud continuum. It will integrate cloud resources offered by multiple stakeholders and allow network functions from different 6G network segments to be composed flexibly and dynamically based on service needs in hybrid cloud environments using orchestration.
.
The following objectives are targeted by 6G-Cloud:
The prime target of 6G-Cloud is to provide the blueprint for the overall service-oriented 6G system architecture design, which is composed of basic network functionalities and multiple control and management frameworks over a multi-stakeholder cloud environment spanning from extreme edge to central clouds. For the first time, mobile system architecture design could be fundamentally different from the previous generation. Following industry needs and technical advances and taking one bold step further, 6G-Cloud will bring E2E service-oriented design to the mobile network and turn the last mile into a truly cloud-native experience.
6G-Cloud targets the operation of service-oriented 6G network functions under complex, multi-stakeholder dynamic cloud environments. To manage the complexity in such an environment, 6G-Cloud will develop three unique frameworks to coordinate cloud-native functions, manage network and cloud resources, and offer intelligence to network functions. 6G-Cloud will design business interfaces for cloud providers that support accounting, service level agreements, security, reliability, trust, and energy efficiency (including powered devices) in a multi-stakeholder dynamic environment.
6G-Cloud will propose a new solution that is based on service-oriented architecture, especially the transfer of the RAN architecture to a service-oriented design, while also considering the challenges and physical limitations. The project will leverage the service-oriented design in the core network and go one major step beyond to integrate two service-oriented domains into an E2E composable and service-based 6G architecture. The proposed integration of RAN and core functions is a key scenario for the 6G-Cloud network service integration capabilities.
The AI/ML framework (AIMLF) will offer unified AI/ML management and orchestration for different segments and frameworks of the 6G-Cloud architecture. The data collection interfaces and functions, AI/ML model management, model training, and distribution will be defined to support flexible and continuous integration/continuous development types of AI/ML deployment in a service-oriented 6G architecture. The AI/ML framework will be able to exploit the network digital twin concept. In the network digital twin workflow, AI/ML mechanisms will be embedded into data collection, federated training, and inference for optimized network observability.
The cloud continuum approach requires a specific approach to 6G management and orchestration. A part of 6G-Cloud is the distributed management and orchestration framework. In the 6G-Cloud case, service orchestration is separated from resource orchestration, which is part of the cloud continuum framework. Moreover, the management and orchestration will allow the AI/ML framework for runtime monitoring and update AI-driven functions used to assist the management and orchestration. The proposed management framework will allow tenants to request deployment, modification, or termination of networks or applications. It will also play a key role in the dynamic integration of different network services.
Leveraging the E2E service-oriented design, 6G-Cloud will develop a unified and open programable control framework to coordinate and customize cloud-native network functions, provision network and cloud resources, and offer interfaces and control mechanisms to integrate with third-party control apps. 6G-Cloud will introduce new mechanisms to build real-time control loops in network functions to unlock fine-grain control programmability. It provides a totally new way to evolve the capability and capacity of network functions and allow the network to evolve continuously and meet service needs.
The KVI of inclusive sustainability, safety, and security, as well as the SDG concerns on service availability, cost efficiency, and accessibility, have strong implications for the system design. 6G-Cloud will conduct the KVI and SDG analysis and reflect these key considerations in the service-oriented system design. A systematic view will be taken to consider these high-level requirements for a holistic solution. The performance and limitations of service-oriented architecture, including energy efficiency, will be intensively studied in the project.
6G-Cloud will develop the prototype of key architecture components. It will build three proof-of-concepts to validate the architecture design from different perspectives and demonstrate the overall project concept based on the selected 6G use cases. The overall system concept will be demonstrated in a final project event.
6G-Cloud targets technologies that will have a clear impact on future standardization activities in various domains, from the management plane to the control plane as well as the service enablement layer. Targeted standardization efforts will be done continuously by project partners who have strong expertise and experience in the field.