The upcoming 6G will represent a complete paradigm shift for global communications. Addressing the critical research verticals toward the envisioned 2030 will require a compelling mix of enabling radio access technologies (RAT) and native softwarized, disaggregated, and intelligent conceptions such as the Open Radio Access Network (O-RAN) architecture. Integrating the Multicast/Broadcast Services (MBS) capability is an appealing feature to overcome the ever-growing traffic demands, disruptive multimedia services, massive connectivity, and low-latency applications. This article discusses the insertion of machine learning (ML)â\euro“based multicasting Radio Resource Management (RRM) solutions in the 6G O-RAN. We review the expected evolution of the MBS capability, including enabling technologies and challenges for the IMT-2030 framework. Moreover, we cover essential aspects at the intersection of MBS, ML-based RRM solutions, and the disaggregated O-RAN architecture, identifying possible scenarios as feature extensions of O-RAN. We present the outcomes of a comprehensive MBS use case simulation oriented to validate our approach, highlighting critical remarks and conclusions.Â

The envisaged fifth-generation (5G) and beyond networks offer unprecedented breakthroughs in media service delivery, stringent requirements, and challenging use cases. In such context, point-to-multipoint (PTM) communications have proved their high efficiency in delivering high-quality multimedia services over wireless networks. For the upcoming 5G system Releases and beyond, the Multicast/Broadcast Services (MBS) capability is an appealing feature to address the ever-growing traffic demands, disruptive multimedia services, massive connectivity, and low-latency applications. This paper addresses the multicast group (MG) channel quality diversity through dynamic single-rate and multi-rate multicasting aided orthogonal multiple access (OMA) and non-orthogonal multiple access (NOMA) over 5G MBS use cases. We characterize and identify conditions and variables interrelation for an effective coexistence of the benchmark conventional multicast scheme (CMS) and subgrouping based on OMA (S-OMA) and subgrouping based on NOMA (S-NOMA). We propose tailored multicast radio resource management (RRM) strategies aided by K-Means subgrouping and particle swarm optimization (PSO). Moreover, we define the interrelation among the variables that shape the performance of the proposed solution and their dynamic behavior in terms of the users’ reception conditions, multimedia service constraints, and network parameters. The presented results are validated under a wide range of users’ channel quality distributions, and network conditions recreated through link-level simulations of a 5G MBS use case operating in millimeter-wave (mmWave).