Machine Learning-Based Cell-Free Support in the O-RAN Architecture: An Innovative Converged Optical-Wireless Solution Toward 6G Networks
John S. Vardakas, Kostas Ramantas, Evgenii Vinogradov, Md. Arifur Rahman, Adam Girycki, Sofie Pollin, Simon Pryor, Philippe Chanclou, Christos Verikoukis
Abstract
One of the most crucial challenges for the global socio-technical domain nowadays is the support of unprecedented use cases for a variety of applications such as augmented reality, 3D holographic telepresence, Industry 4.0, robotics, e-health, and pervasive connectivity. In this environment, beyond 5G intelligent networks that are able to provide enhanced flexibility through the dynamic allocation of the system's resources, while realizing the perceived zero-latency vision, are envisioned to ensure a smooth transition toward 6G. The proposed innovative converged optical-wireless network configuration targets to enable such a transformation, through the development of novel radio access networking solutions. These advanced radio-edge architectures are designed by utilizing the distributed cell-free concept and the serial fronthaul approach, while contributing innovative functionalities to the O-RAN project. At the same time, the proposed approach facilitates integrated connectivity of both mobile and fixed services, which share the same edge and midhaul infrastructures, while both are served by a core. The proposed network architecture can be envisioned as the key technology enabler to satisfy the requirements of future 6G networks through the modifications of the O-RAN interfaces toward a cell-free-enabled solution, and through the incorporation of novel Machine Learning techniques at the Radio Edge.