Performance Analysis of Multi-User Task Offloading in Mobile Edge Computing Networks: A Hierarchical Modeling Approach

Recently, the explosive growth of connected intelligent user devices has generated massive amounts of latency sensitive and computationally intensive applications and services, which leads to a sharp increase in network traffic. Therefore, UAV-assisted mobile edge computing networks (UMECN) are proposed to alleviate network congestion and reduce response time. This paper focuses on quantitatively evaluating UMECN service performance under different parameters, which helps to identify performance bottlenecks, and provide a basis for optimizing services, thereby improving service quality, enhancing user experience. Firstly, we propose the hierarchical model for capturing the dynamical behaviors of UMECN system, supporting interruption/resume uploading and returning process of delay sensitive tasks due to UAV movement. And then, we derive the performance metric formulas of UAV movement and UMECN system, and propose a multi-factor selection strategy base the task historical waiting time, queue length, and the number of remaining resources (WQRSS). Finally, we verify the approximate accuracy of the proposed formulas and model by comparing numerical and simulation results, demonstrate the effectiveness of the model through comparative experiments, and evaluate the impact of different parameters on UMECN service performance, which can help service providers effectively manage UMECN system.