A Service-Oriented Optimization Framework for Edge Caching With Revenue Maximization and QoS Guarantees
Chia‐Cheng Hu, Jiao-Yan Zeng · IEEE Transactions on Services Computing · 2025
The rapid proliferation of mobile applications and data-intensive services, such as augmented reality and real-time analytics, necessitates efficient content delivery mechanisms in Mobile Edge Computing (MEC) environments. MEC enhances service responsiveness by caching content closer to end users; however, the constrained storage capacities of edge servers pose challenges in maintaining optimal Quality of Service (QoS). This paper presents a novel service-oriented content caching framework that optimizes resource allocation and revenue generation while ensuring QoS compliance. We introduce a dynamic fee-based pricing model that adapts service charges based on content retrieval latency, incentivizing improved service performance. The caching optimization problem is formulated as an Integer Linear Programming (ILP) model, and a computationally efficient approximation algorithm leveraging Linear Programming (LP) relaxation and rounding techniques is proposed to derive near-optimal solutions. Additionally, a resource expansion model is integrated to dynamically extend storage capacity in response to evolving service demands, ensuring scalable and cost-effective content provisioning. Extensive theoretical analysis and simulations validate the proposed approach, demonstrating a 22.4% increase in total service revenue and a 31.7% reduction in average content access delay compared to baseline strategies. This work contributes to services computing by providing a mathematically rigorous and computationally efficient framework for dynamic content management in edge networks.