Research on Computation Offloading Strategy Based on Quality of Experience Optimization in Mobile Edge Computing

With the widespread adoption of mobile internet and intelligent terminal devices, the emergence of increasingly computation-intensive and latency-sensitive applications poses significant challenges to resource-constrained mobile devices. As a promising computational paradigm, Mobile Edge Computing (MEC) offers an effective solution for reducing latency and energy consumption by offloading tasks to proximate edge servers. However, in multi-user scenarios, the heterogeneous nature of task demands and dynamic competition for edge resources render the collaborative optimization of offloading strategies—aimed at ensuring user experience quality—particularly challenging. To address these issues, this paper proposes a distributed intelligent offloading scheme based on the Dueling Double Deep Q-Network algorithm. The proposed scheme seeks to maximize the overall user experience quality of the system by jointly optimizing average latency, average energy consumption, and task offloading success rate. Simulation results comprehensively validate the superiority of the proposed approach, demonstrating its effectiveness in significantly enhancing user experience quality.