An Innovative Multi-Scale Strategy-Based Decision Engine for Zero-Touch Management and Orchestration in 6G

In the upcoming era of 6G, mobile networks will need to handle demanding applications such as extended reality, digital twins, holographic telepresence, and immersive communication. They will also have to meet stricter application requirements across the edge cloud continuum. These new applications will raise expectations for performance, reliability, ubiquity, trustworthiness, security, openness, and sustainability that will drive innovation and bring about transformative changes across the architecture of future mobile networks. Zero-touch management will become a necessity for delivering the next generation of services in 6G networks with speed and agility, enabling network self-configuration, selfoptimization, self-monitoring, and self-healing without further human intervention at both local and global scopes. Decision Engines (DEs), leveraging AI, will play a critical role in the orchestration and management of 6G networks by optimizing network performance and user experience through real-time, automated, and intelligent decisions. An innovative DE is proposed and empowered by Hierarchical Reinforcement Learning (HRL), a multi-scale strategy that orchestrates decisions across all network domains through local agents. The architectural and algorithmic assumptions of this DE will ensure efficient, scalable, and autonomous network operations, enhancing not only the performance and reliability of 6G networks but also paving the way for new applications and services that can leverage the full potential of next-generation connectivity.