MAIN/OBJECTIVES

OBJECTIVE 1. Design and demonstrate scalable, distributed cell-free massive-MIMO networks supporting massive AP deployments.

In FREE6G-RadEdge, we aim to design the architecture and protocols to provide scalable cell-free networks that offer best-in-class connectivity and latency while maximally exploiting all available distributed resources. Optical and mmWave point-to-point (PtP) and point-to-multipoint (PtMP) wireless links are all exploited, and all distributed processing facilities are used efficiently with minimal cooperation overhead.

OBJECTIVE 2. Design and implement a cell-free vRAN for 6G, aligned with the O-RAN Alliance architecture.

The FREE6G-RadEdge project will be aligned with the O-RAN Alliance architecture, which disaggregates the RAN in a number of virtualized RAN (vRAN) Network Functions (i.e., vO_DU, vO_CU) and complements 3GPP standards with new packet-based interfaces. FREE6G-RadEdge aims to deliver cell-free support for O-RAN for the first time, implementing a foundation of O-RAN compliant vO_CU and vO_DU elements, designing and implementing cell-free MAC, RRM and PHY sublayers. Moreover, appropriate modifications to control plane protocols and the O-RAN specified interfaces (i.e., E2, O1) and Near-RT RIC will be proposed, to support practical cell-free operation with fully distributed processing.

OBJECTIVE 3: Architect a disaggregated SDN control plane towards Fixed-Mobile Convergence.

Fixed-Mobile Convergence (FMC) is a key direction of 6G networks, aiming to converge fixed and mobile networking at the edge of the network, served by a common 6G core. FREE6G-RadEdge aims to converge the RAN and Midhaul network segments with existing PtMP fibre deployments (e.g., for FTTH) under a hierarchy of domain-specific SDN controllers, that can react to workload variations in near-real time. Thus, increased statistical multiplexing through load balancing of pooled networking resources can be achieved, with significant CAPEX/OPEX savings. Moreover, FREE6G-RadEdge will support unified orchestration for fixed and mobile services.

OBJECTIVE 4: Deployment of an Elastic Edge Computing paradigm with Cloud-Native technologies.

FREE6G-RegEdge envisages a new paradigm of Edge infrastructures based on the notion of Elastic Edge Computing, aiming to overcome the isolation and underutilization of resources deployed at Edge nodes, and offering zero perceived latency to smart connectivity applications. To this end, FREE6G-RegEdge will propose innovative extensions to the MEC system to facilitate the disaggregation of Cloud-Native MEC applications both vertically (across may tiers of Edge nodes) and horizontally (i.e., across the same tier) aiming for the optimal distribution of latency budgets.

OBJECTIVE 5: Deliver a Self-driven infrastructure with pervasive, ML-driven control.

6G networks will be characterized by Self-Driven infrastructures, with pervasive ML and closed-loop autonomy at all layers. However, since a “global brain” that controls the infrastructure top-down is not feasible for scalability purposes, FREE6G-RegEdge aims to deliver Self-Driven autonomy with a novel, distributed approach that involves Analytic Engines at all tiers of the Edge infrastructure, and Decision Engines at the two Core-Tier orchestration subsystems (i.e., the NFVO, and the Multi-access edge orchestrator).

OBJECTIVE 6: Policy-driven security, privacy, and trust in multi-tenant infrastructures.

Multi-tenancy is foreseen as one of the main pillars of 6G. FREE6G aims to support the decentralized collaboration, via smart contracts, of non-trusted tenants without them having to share their business or operational data. Furthermore, FREE6G aims to protect user data from rogue tenants in perpetuity with an advanced policy-driven framework, and with distributed ML-driven Threat Detection and Threat Analysis Engines.

OBJECTIVE 7: implementation and proof-of-concept of the FREE6G solutions.

FREE6G will carry out a series of demonstrators as an outcome of the proof-of-concept (PoC) phase to demonstrate the ability of all its targeted network technologies.

OBJECTIVE 8: Dissemination, standardization, and exploitation of FREE6G

Dissemination, standardization, and exploitation of the FREE6G-RadEdge technologies. Special focus is given to key standardization bodies to sustain standardization in the telecom and media sectors.

• Machine learning-based cell-free networks for 6G – Radio Edge (FREE6G-RadEdge): Focuses on innovations at the RAN and fronthaul domains exploiting the potential of cell-free networking in future 6G networks. Combines mmwave hybrid beamforming radios, distributed processing, algorithms for adaptive access point (AP) clustering and open disaggregated RAN initiatives for cost-effective deployment. At the midhaul interfaces, optical ethernet technologies will be leveraged and the network will allow flexible sharing of the optical fibres with fixed services.

• Machine learning-based cell-free networks for 6G – Regional Edge (FREE6G-RegEdge): Provides elastic edge Cloud with machine learning (ML) driven dynamic slicing support and zero-perceived latency for the mobile edge computing (MEC) applications. This implies supporting advanced intra and inter data center load balancing to reduce the latency. Dynamic slicing approaches will be explored to trigger predictive slice reconfiguration.

• Machine learning-based cell-free networks for 6G – Security (FREE6G-Security): Targets the security and privacy of multi-tenant RAN infrastructures. Blockchain-based solutions fostering collaboration between tenants via smart contracts, and supporting the flexible negotiation of network slice contracts, will be developed. Finally, federated learning solutions with blockchain technologies will be employed to provide system-level security in 6G server-less architectures.