🤖 AI Summary
To address bandwidth constraints, heterogeneous user demands, and time-varying wireless channels in short-video multicast delivery, this paper proposes a QoE-driven dynamic multicast resource management framework. We design a lightweight online scheduling mechanism that jointly optimizes bandwidth allocation, multicast group formation, and startup latency—integrating short-video viewing behavior modeling with real-time channel state awareness for the first time. The problem is formulated as an integer nonlinear program, and solved via a QoE-aware multi-objective online heuristic algorithm, overcoming limitations of conventional static multicast resource provisioning. Experimental results demonstrate that, compared to baseline schemes, our approach reduces bandwidth consumption by 37.2%, decreases initial-frame latency by 51.8%, and achieves a Mean Opinion Score (MOS) exceeding 4.1—representing a 29.4% improvement in user satisfaction. Furthermore, the framework exhibits strong scalability and robustness under LTE/5G hybrid network topologies.
📝 Abstract
The surge in popularity of short-form video content, particularly through platforms like TikTok and Instagram, has led to an exponential increase in data traffic, presenting significant challenges in network resource management. Traditional unicast streaming methods, while straightforward, are inefficient in scenarios where videos need to be delivered to a large number of users simultaneously. Multicast streaming, which sends a single stream to multiple users, can drastically reduce the required bandwidth, yet it introduces complexities in resource allocation, especially in wireless environments where bandwidth is limited and user demands are heterogeneous. This paper introduces a novel multicast resource management framework tailored for the efficient distribution of short-form video content. The proposed framework dynamically optimizes resource allocation to enhance Quality of Service (QoS) and Quality of Experience (QoE) for multiple users, balancing the trade-offs between cost, efficiency, and user satisfaction. We implement a series of optimization algorithms that account for diverse network conditions and user requirements, ensuring optimal service delivery across varying network topologies. Experimental results demonstrate that our framework can effectively reduce bandwidth usage and decrease video startup delay compared to traditional multicast approaches, significantly improving overall user satisfaction. This study not only advances the understanding of multicast streaming dynamics but also provides practical insights into scalable and efficient video distribution strategies in congested network environments.