Leveraging User Perception for 6G Edge-Cloud Orchestration of Networked Extended Reality

Abstract

As extended reality (XR) technologies evolve, optimizing quality of experience (QoE) for end-users becomes increasingly challenging due to the stringent network requirements of these applications. Traditional quality of service (QoS) metrics often fall short, making user perception central to effective network management. This challenge becomes even more pronounced with the advent of 6G networks, where XR applications will be deployed on distributed computing resources across the compute continuum (CC). This article presents our vision for edge-cloud orchestration in 6G, featuring a hybrid QoE model that integrates network and physiological metrics with subjective user feedback. In addition, we leverage reinforcement learning (RL) to efficiently manage QoE across edge-cloud infrastructures within the CC. By addressing the dynamic and complex nature of XR applications from both technical and human perspectives, our approach aims to meet the stringent low-latency requirements of these immersive applications. The article also highlights recent advancements in QoE modeling through a collaborative virtual reality (VR) use case, demonstrating the importance of integrating objective and subjective metrics for effective QoE management. This article aims to provide a foundational framework for QoE management in 6G networks, paving the way for future research and innovation in this rapidly evolving domain.