Partial Offloading with Stable Equilibrium in Fog-cloud Environments using Replicator Dynamics of Evolutionary Game Theory

In this research, we propose a solution for partial offloading based on the replicator equation of evolutionary game theory. The ultimate goal is for users to be in a stable equilibrium based on their priorities in terms of energy consumption, latency, and computational cost so that no user is willing to change his/her strategy. The results of solving the replicator equation followed by statistical analysis show that the proposed method has a remarkable performance improvement compared to the state-of-the-art methods. Our method not only shows a better average based on the weighted combination of energy consumption, latency, and cost but also has a lower variance compared to the full local execution and the full offloading methods. Our results also show that when the number of users is very large, the system can reach equilibrium with the least amount of information exchange among players.本研究基于进化博弈论的复制者方程（replicator equation），提出了一种面向部分卸载的解决方案。本研究的最终目标是让用户基于能耗、时延与计算成本的优先级达成稳定均衡，使所有用户均无意愿变更自身策略。通过求解复制者方程并开展统计分析后的结果表明，相较于现有前沿方法，所提方案的性能提升显著。相较于完全本地执行与完全卸载方案，所提方法不仅在能耗、时延与成本的加权组合指标上表现更优，同时方差更低。研究结果同时显示，当用户规模较大时，系统可通过参与者间最少的信息交互达成均衡。