Optimization of task unloading strategy based on Game Theory in cloud edge collaborative system
Zhongbo Hu, Suzhen Wang, Lisha Hu, Yongchen Deng · 2022
Task unloading can reduce task completion time and save system energy consumption by migrating processing tasks on terminal devices to edge servers, which is the key research content of edge computing. At present, most of the research on edge computing task unloading is edge end centralized architecture, without considering the powerful computing power and resources of cloud center. In order to make full use of cloud and edge server resources, this paper proposes an optimization scheme of coarse-grained computing task unloading strategy under cloud edge cooperation. Considering the heterogeneity of mobile devices and the interference between channels, in order to minimize the system delay and energy consumption, the task unloading decision of multiple rational end users is established as a task migration model based on game theory, and an improved particle swarm optimization algorithm is proposed to obtain the optimal task unloading strategy set. Simulation results show that the proposed scheme can significantly reduce the system overhead cost, and can expand the scale with the increase of the number of mobile devices.