Project title: 5G Network Atchitecture towards 6G Networks
MCNS is participating in a research project with technology knowledge transfer in ETSI in Sophia-Antipolis, France on the 5G network cloudification in the forthcoming 6G era.
6G will evolve from 5G by leveraging cloudification and the flexible architecture of 5G Core (5GC). It will expand network virtualization and edge computing to deliver ultra-low latency and massive scalability. By integrating AI/ML, 6G will enable intelligent automation, real-time analytics, and seamless orchestration across devices and services. Enhanced slicing and non-terrestrial networks will provide ubiquitous, reliable connectivity. This evolution will foster immersive experiences and support advanced applications like pervasive intelligence [1]
Key Objectives and Considerations
In June 2023, ITU-R Working Party 5D (WP5D) finalized the 6G Framework outlined in ITU-R M.2160, marking a significant step forward in next-generation communication standards. This framework builds upon the foundation of IMT-2020 (5G) by introducing expanded and innovative usage scenarios to address emerging technological and societal needs. It also defines enhanced and entirely new capabilities to support advanced applications, such as immersive communication, ubiquitous intelligence, and sustainable connectivity. The 6G Framework aims to drive innovation by pushing the boundaries of wireless technologies and setting the stage for seamless integration of terrestrial, non-terrestrial, and AI-driven networks, fostering a truly connected and intelligent world.
Figure 1. 3GPP RAN1 Release 19 5G to 6G evolution tasks vs timeline
3GPP publishes standard technical specifications, for which a release number is assigned to each functional set. After formulating Rel-15, 3GPP issued Rel-16 and Rel-17 to expand the scope of functionality and improve the performance of 5G. 3GPP calls Rel-18 and follow-up releases “5G-Advanced” and started the development of the Rel-18 specifications in 2022. 5G-Advanced is targeted for commercial service in the late 2020s and will serve as a steppingstone toward 6G, which is targeted for commercial service by around 2030.
3GPP has recently completed the Rel18 normative specifications regarding service and operational requirements to support network slicing and work has started on both system architecture aspects and related management and orchestration capabilities. It defines the enablers to support Network Slicing involving 5QI quality indicators, and L1 / L2 configurations, with possibility to prioritize some services or devices in lower layers (e.g. at MAC scheduler level).
Figure 2. 3GPP TSG RAN Release 19 5GC and slicing enhancements
Rel-18 was the first release in which Artificial Intelligence (AI) and Machine Learning (ML) was studied in RAN1 and despite the potential for a large amount of RAN1 work, the study focused on only three representative use cases: CSI feedback, beam management, and positioning. For each use case, RAN1 studied the benefits of augmenting the air interface with features enabling AI/ML based algorithms for enhancing performance and/or reducing overhead. In addition, extensive study was done on the AI/ML framework that includes defining stages of AI/ML related algorithms, various levels of collaboration between UE and gNB, lifecycle management, and necessary dataset(s). The RAN1 study was extended for an extra quarter to ensure proper study on all three Rel-18 use cases. The work will lay the foundation for future normative work on AI/ML and for additional studies on other use cases.
The advent of cloud-native design has revolutionized network deployment, introducing advanced implementation technologies and enhanced methods like streamlined software lifecycle management. This innovation paved the way for the 5G Core (5GC) network’s Service-Based Architecture (SBA), as standardized by 3GPP. Unlike traditional architectures relying on point-to-point protocols, SBA employs cloud-optimized service-based interfaces (SBIs), fostering a dynamic and scalable framework.
SBA’s flexibility enables seamless integration of new network functions (NFs), exemplified by the growth from 22 NFs in 3GPP Release 15 to 45 in Release 17. This evolution reflects the expanding scope of 5GC functionalities, driven by the need to support diverse features in modern networks. The number of SBIs has surged even more significantly, surpassing 110 in Release 17, highlighting the architecture’s adaptability to emerging technological demands.
By leveraging SBA, the 5GC achieves unparalleled extensibility, positioning itself as a cornerstone for future innovations. This approach ensures networks remain agile and capable of meeting the evolving needs of 5G and beyond [2].
Figure 3. 3GPP Rel-19 5G cloud core architecture evolution towards 6G
Project Conclusion
Discussions have been also related to the 3GPP TSG RAN plenary meeting for 6G initiative.
During the 3GPP RAN#106 meeting in Madrid on December 12th, a pivotal 6G study item was approved, marking a major step in advancing global 6G standardization. This initiative focuses on collaborating with ITU to define technical performance requirements (TPRs) and explore deployment scenarios, requirements, and potential directions for 6G radio access technologies. The study, detailed in RP-243327, will evaluate minimum TPRs based on ITU-R M.2160 recommendations, along with their target values and assumptions. Its findings will be shared with ITU-R WP5D as a foundational reference for further 6G studies within 3GPP [3].
Customer related Web links:
[1] https://www.etsi.org/technologies/nfv
[2] https://www.ericsson.com/4904ea/assets/local/reports-papers/ericsson-technology-review/docs/2023/6g-network-architecture.pdf
[3] https://www.3gpp.org/news-events/3gpp-news/ran-6g-study1
