Recent Advances in Deep Learning with Optimization-Based Task Scheduling and Computing Resource Allocation for VR Video Services in Advanced 6G Networks: A Systematic Review

Abstract

The emergence of 6G networks is expected to revolutionize immersive applications such as Virtual Reality (VR) video services, which demand ultra-low latency, high bandwidth, and efficient resource management. However, VR applications generate massive data streams and require real-time processing, making task scheduling and computing resource allocation critical challenges. Traditional optimization approaches fail to handle the dynamic and stochastic nature of 6G environments, necessitating the integration of deep learning with advanced optimization techniques. This systematic review explores recent advances in deep learning-based and optimization-driven frameworks for joint task scheduling and resource allocation in VR-enabled 6G networks. Deep Reinforcement Learning (DRL) has emerged as a key technique for dynamic decision-making in edge computing environments, enabling intelligent task offloading and adaptive resource allocation.  Furthermore, Lyapunov optimization-based methods provide theoretical guarantees for system stability and latency minimization by transforming complex optimization problems into queue stability models.  Recent research also focuses on hybrid frameworks that combine DRL with Lyapunov optimization to achieve efficient scheduling and resource allocation in highly dynamic networks. These approaches enable real-time adaptation, reduced latency, and improved Quality of Experience (QoE) for VR applications.  This review analyses recent studies (2020–2023), compares methodologies, and identifies research gaps such as computational complexity, scalability, and energy efficiency. The findings highlight that hybrid deep learning and optimization-based approaches are the most promising solutions for next-generation 6G VR systems.