ENTWINE-6G · End-to-end Network digital TWIns for Native AI drivEn 6G architectures

Lines connect the coordinator with its partners.

Project objective

6G research is already underway to shape the new generation telecommunication networks, aiming to deliver fully autonomousmulti-technology, potentially 3D networking infrastructures. One of the key enablers of 6G is undoubtedly the Native integration ofAI/ML at all system levels, components, and technology domains. However, this approach suffers in terms of decreasedtrustworthiness, which acts as a barrier to the adoption of data-driven technologies in future 6G networks. To this end, ENTWINE-6Gresearch project aims to propose, design, and validate a novel automation architecture, aiming for the integration of Network DigitalTwins (NDTs) into an AI-driven 6G System, to facilitate trustworthy automation of multi-technology 6G infrastructures. ENTWINE-6Gproposes the cooperation of low-fidelity NDTs at the upper layers of the 6G architecture with high-fidelity twins that accurately model6G infrastructure components. Moreover, ENTWINE-6G will contribute Trustworthy AI techniques, along with their associatedenabling technologies and theoretical frameworks that will be leveraged to increase robustness and ensure SLA compliance. Finally,Generative techniques, including neural rendering, 3D diffusion, and NeRF, will emulate complex spatial and environmentalconditions that will be validated with powerful digital twins tools that support Ray-Traced channels.

Original text from CORDIS.

Participants

• ATHINA-EREVNITIKO KENTRO KAINOTOMIAS STIS TECHNOLOGIES TIS PLIROFORIAS, TON EPIKOINONION KAI TIS GNOSIS · MAROUSSICoordinatorGreece
• EURECOM GIE · BiotFrance
• FUNDACIO PER A LA UNIVERSITAT OBERTA DE CATALUNYA · BARCELONASpain
• IQUADRAT INFORMATICA SL · BarcelonaSpain
• NVIDIA DENMARK APS · RoskildeDenmark
• SORBONNE UNIVERSITE · ParisFrance
• TELEFONICA INNOVACION DIGITAL SL · MADRIDSpain
• UNIVERSIDAD POLITECNICA DE MADRID · MadridSpain
• UNIVERSITAET BREMEN · BremenGermany
• UNIVERSITAT POLITECNICA DE CATALUNYA · BARCELONASpain

Links

• View on CORDIS
• DOI: 10.3030/101312201