基于博弈论和粒子群优化的移动边缘计算任务卸载方法

Game Theory and Particle Swarm Optimization Based Task Offloading Method in Mobile-Edge Computing

移动边缘计算(mobile-edge computing,MEC)是一种新兴的计算范式,移动设备可以通过将计算密集型任务卸载到边缘服务器上来降低本地计算能耗和计算时延。首先,该文研究了在微蜂窝基站密集区域场景下的多移动设备独立任务集计算卸载问题,其中每个微蜂窝基站配备了一个计算性能有限的MEC服务器。为了尽可能地降低移动设备的任务集计算能耗和计算时延,使用博弈论的方法将该问题建模为一个非合作多移动设备计算卸载策略博弈。通过对该博弈的分析,证明了其纳什均衡的存在性和有限改进性。然后,设计了一个基于博弈论的分布式计算卸载算法(game theory based distributed computation offloading algorithm,GDCOA),并在GDCOA中引入了一个基于粒子群优化(particle swarm optimization,PSO)的移动设备任务集卸载策略改进算法(PSO based improving computation offloading policy algorithm,PSOIPA)。GDCOA在有限次迭代后可以达到一个均衡状态。最后,通过仿真对比实验证明了本文所提出的算法GDCOA可以获得较好的计算卸载性能。

Mobile-edge computing(MEC)is an innovative computing paradigm.Mobile devices can reduce local computation energy consumption and delay by offloading computation intensive tasks to the edge servers.In this paper,we first study the computation offloading problem for multiple mobile devices with independent task sets in the dense area of microcell base stations,where each microcell base station is equipped with a computationally limited MEC server.To reduce the task sets computation energy consumption and delay of the mobile devices as much as possible,adopting a game theoretic approach,the problem is formulated as a non-cooperative multi-mobile-device computation offloading strategy game.Through analysis,the Nash equilibrium existence and the finite improvement property of the game are proved.Then,we design a game theory based distributed computation offloading algorithm,namely GDCOA,and introduce a particle swarm optimization(PSO)based improving computation offloading policy algorithm named PSOIPA in it.GDCOA can reach an equilibrium state after a finite number of iterations.Finally,the simulation and comparison experiments corroborate that the proposed algorithm GDCOA in this paper can achieve better computation offloading performance.