Concept and Methodology
Overall Concept
As the world moves from the 5G towards the 6G era, the mobile communications fabric needs to be architected differently to accommodate the emerging stringent requirements of innovative extreme future-looking applications that cannot be served by existing 5G mobile networks. Heading towards the next decade, when 6G is expected to be widely deployed, 5G application types will be redefined by morphing the classical service classes of Ultra-Reliable and Low-Latency Communications (URLLC), enhanced Mobile BroadBand (eMBB), and massive Machine-Type Communications (mMTC) and introducing new services.
ADROIT6G proposes disruptive innovations in the architecture of emerging 6G mobile networks that will make fundamental changes to the way networks are designed, implemented, operated, and maintained. Such innovations include:
AI/ML-powered optimizations across the entire network, for high performance and automation.
Transforming the cellular network to a fully cloud-native network software, which can be implemented across a variety of edge-cloud platforms, including Non-Terrestrial Networks, with security built integrally into the network user plane.
Software-driven, zero-touch operations and ultimately automation of every aspect of the network and the services it delivers.
ADROIT6G innovations, functionalities, and performance will be validated through 3 representative extreme 6G use cases (UCs), namely the holographic telepresence, Industrial IoT, and collaborative robots/drones in corresponding Proof of Concepts over a number of well-established 5G testbeds, which will be upgraded to support ADROIT6G innovations and architectural elements.
ADROIT6G Methodology
ADROIT6G aims to explore and integrate disruptive, best-of-breed technologies to produce a truly novel system. The project is addressing a radically new approach for a next-generation AI-enabled open architecture for future 6G networks, and therefore, it requires continuous design and refactoring. The design, implementation, testing, and usage of ADROIT6G will go in parallel with the validation in lab settings, and a lot of feedback and re-design of platform components may be expected because of a better understanding of the challenges faced with the integration of novel concepts, components, and their usage in future looking applications. Henceforth, new technologies and tools may succeed and be integrated into the design/development cycle of the project. ADROIT6G will be open to rapid changes and elicitation of new requirements stemming from the UCs in the PoCs.
ADROIT6G will be executed over contiguous and sequentially aligned phases focusing on the requirements’ analysis, the definition of UCs scenarios and methodologies, the design of the architecture, the development of the system’s building blocks and their integration, technology validation in labs, and impact maximization activities. These phases aim to provide and improve a consolidated set of results, and therefore, the following approaches will be used to enable faster software development:
Service-oriented methodology for the design approach.
An agile approach for the development of the ADROIT6G software components.
Machine Learning Operations (MLOps) methodology for the development of the ADROIT6G AI/Machine Learning (ML) algorithms.
Theoretical considerations, simulation, emulation and upgraded 5G testbeds for the validation of ADROIT6G concept and technologies.