EMPOWER CONVERGED OPTICAL WIRELESS CONFIGURATIONS WITH CELL-FREE TECHNOLOGIES FOR HIGH-DENSITY 6G NETWORKS

Informazioni relative al progetto

EMPOWER-6G

ID dell’accordo di sovvenzione: 101120332

DOI

10.3030/101120332

Data della firma CE 14 Luglio 2023

Data di avvio 1 Ottobre 2023

Data di completamento 30 Settembre 2027

Finanziato da

• Marie Skłodowska-Curie Actions (MSCA)

Costo totale

Nessun dato

Contributo UE

€ 3 821 918,40

Coordinato da

IQUADRAT INFORMATICA SL
Spain

Periodic Reporting for period 1 - EMPOWER-6G (EMPOWER CONVERGED OPTICAL WIRELESS CONFIGURATIONS WITH CELL-FREE TECHNOLOGIES FOR HIGH-DENSITY 6G NETWORKS)

Periodo di rendicontazione: 2023-10-01 al 2025-09-30

Sintesi del contesto e degli obiettivi generali del progetto

EMPOWER-6G proposes a break-through converged optical-wireless architecture that makes reality an efficient 6G Cell-Free (CF)-based access network for enabling the requested 1 Tb/s peak data rate and deployments with over 10 million devices/km2 and for supporting new advanced user-services (e.g. holographic telepresence, advanced AR/VR, massive digital twinning, flying networks, etc.). The EMPOWER-6G efforts are channeled towards three ambitious objectives, corresponding to three inter-disciplinary but inter-connected research areas that are further decomposed to 9 specific research challenges:

At the radio edge network level, EMPOWER-6G aims to develop:
1. a scalable CF architecture that offers best-in-class Spectral Efficiency (SE) (within 1dB of the optimum), improved throughput, security, and latency, while maximally exploiting all available distributed resources;
2. a flexible wireless point-to-multipoint mmWave fronthaul solution based on novel hybrid beamforming and beam multicasting schemes, achieving at least double SE due to network densification,
3. a virtualized CF RAN that is aligned with the O-RAN alliance, by considering joint radio and computing resource optimization at the radio edge network, thus achieving at least double per-user UL throughput compared to traditional CF networks

At the regional-edge network level, EMPOWER-6G aims to develop:
1. an optical regional edge network with FTTH support with a significant reduction of the fronthaul cost,
2. regional network configurations based on optimum functional splitting options, that maximize energy efficiency,
3. control-plane protocols at the regional edge for CF networking support that are aligned with the O-RAN, that achieve 20% OPEX reduction

At the network management level, EMPOWER-6G aims to develop solutions for:
1.the orchestration and management of multiple DUs through the design of an efficient resource scheduler, targeting improved user data-rate by a factor of 3,
2. SDN-enabled intelligent traffic management methods and ML-based slice reconfiguration mechanisms, targeting at least 20% energy savings via traffic aggregation;
3. a self-driven virtual elastic infrastructure, based on a novel ETSI-compliant MEC platform and a blockchain-enabled multi-tenant orchestration system¸ targeting zero perceived latency due to multi-domain disaggregation of MEC apps

Lavoro eseguito dall’inizio del progetto fino alla fine del periodo coperto dalla relazione e principali risultati finora ottenuti

During the first period of the project,the following main activities were performed:
• the EMPOWER-6G website is up and accessible.
• the project's data management plan has been defined
• execution of EMPOWER-6G's school 1, 2, 3, 5
• a video prepared by the project’s DCs and disseminated
• Organization of frequent meetings among beneficiaries and the DCs
• Continuous updates of the project’s website and social media accounts with project’s news, updates, etc.
• Green open access for all scientific publications of the project
• Submission of all relevant deliverables

At the technical level, during the second reported period the DCs worked on the development of the beyond state-of-the-art approaches, as it was designed during the first reporting period; The new developments are listed in Deliverables D4.1 D5.1 D6.1 while they are presented in papers published in journals and conference proceedings.

Progressi oltre lo stato dell’arte e potenziale impatto previsto (incluso l’impatto socioeconomico e le implicazioni sociali più ampie del progetto fino ad ora)

The beyond state-of-the-art research contributions of EMPOWER-6G were obtained through the 3 technical WPs (WP4, WP5, WP6):
WP4: Radio-edge infrastructures for 6G networks. The beyond state-of-the-art contributions for the first reported period are listed below:
• exploring stochastic-based AP switching strategies for improving the energy efficiency of cell-free MIMO systems,
• developing AI-based precoding mechanisms for distributed MIMO systems
• developing a novel Reinforcement Learning (RL)-driven framework for adaptive computational load offloading in converged Software-Defined Networking (SDN) control planes, specifically addressing the ultra-low latency and high-capacity requirements of 6G
• designing and developing a novel O-RAN compatible CF RAN architecture with integrated Physical Layer Security (PLS)

WP5: Regional-edge infrastructures for 6G networks. The beyond state-of-the-art contributions for the first reported period are listed below:
• designing single-fiber optical networks for bi-di communication,
• designing a a user-centric, intelligent clustered approach for delivers high spectral efficiency cell-free systems
• developing an analytical software tool to determine optimal transmission using XR optics combined with an interlacer filter, with special attention on the correctness and speed of the model.
• developing a first solution for a hierarchical SDN control plane solution that disaggregates control functions to Regional Edge nodes, enabling sub-second decision-making enhanced by machine learning-based traffic prediction and decentralised coordination across network tiers
• developing a network simulation platform capable of emulating realistic environments, simulating the ORAN framework, for testing

WP6: 6G network management and control. The beyond state-of-the-art contributions for the first reported period are listed below:
• developing a a seamless handover technique which utilizes the inter DU interface in O-RANs
• developing a a multi-agent RL framework that jointly allocates uplink power, offload ratio, and MEC CPU shares under deadline constraints, trained with dynamic traffic and variable CU compute to emulate slicing
• developing solutions for elastic resource allocation to address the scalability and flexibility required by edge computing environments
• developing a a FastAPI-based decision engine with a background anomaly detection loop, packaged with ML libraries for future algorithmic integration
• developing a unified framework for dynamic and adaptive SLA decomposition across multi-domain network architectures, targeting the RAN, transmission-network, and core-network layers.

The scientific outcomes of the three technical WPs have already produced 1 journal and 4 conference publications. This number will be significantly increased in the second period of the project.