Research On Resource Allocation Strategies Based On Game Theory In Mobile Edge Computing

With the rapid development of various information technologies,computationally intensive applications such as smart cities,autonomous driving,and virtual reality are increasingly appearing in daily life.These applications have high requirements for data transmission rate and data processing capability,but they are distributed on various mobile devices,such as mobile phones,smart watches,vehicles,etc.The computing power of these devices is difficult to complete computing intensive tasks on time and obtain the required results,which greatly affects the user experience.Therefore,mobile edge computing(MEC),a new computing mode,has emerged.In mobile edge computing,it is allowed to set up an edge server closer to the mobile device.The edge server has stronger computing power than the mobile device.The computing intensive tasks on the mobile device can be offloaded to the edge server for computing,thus completing computing tasks more efficiently and effectively improving the user experience.Because there are many mobile devices and edge servers in the mobile edge computing network,it must involve how to allocate computing power effectively,so that the network efficiency is higher and users can get a better experience.However,edge servers may come from different operators,so there may be competition for tasks between different edge servers.Game theory is an effective tool to discuss the competition and cooperation between different individuals.Therefore,this paper intends to use game theory tools to discuss the resource allocation strategy in mobile edge computing.The innovative work of this article mainly includes:Firstly,for static scenarios,where edge servers serve the same mobile device for a long time,such as IoT application scenarios,large VR game centers,etc.In this scenario,this article proposes a bidirectional update strategy based on potential games,while optimizing the efficiency and task completion rate of edge servers.In this strategy,a potential game model was constructed for the task competition relationship between edge servers and proved to be solved.Then,corresponding solving algorithms were designed based on the characteristics of the potential game model to maximize the efficiency of edge servers.At the same time,a reverse update loss function is designed,and the online learning algorithm is used to optimize the overall task completion rate of MEC system.The simulation results show that the proposed algorithm can effectively improve the efficiency and task completion rate of edge servers compared to other benchmark algorithms.Secondly,for dynamic scenarios,that is,scenarios where mobile devices have strong mobility but are privacy sensitive,such as mobile intelligent device intensive scenarios.In this scenario,mobile devices have strong mobility and frequent task assignments,resulting in task queues on edge servers,posing a challenge to system stability.This paper proposes an online game theory strategy based on Lyapunov.While considering the benefits of edge servers,task completion rate and energy consumption of mobile devices,it needs to consider maintaining system stability.Analyze the stability of task queues in edge servers using Lyapunov optimization and construct corresponding strategies for solution.The simulation results show that compared to other benchmark algorithms,the proposed algorithm can effectively improve the efficiency and task completion rate of edge servers while ensuring system stability,and reduce mobile device energy consumption.Thirdly,for online caching scenarios,such as car networking,where mobile devices have strong mobility and are not sensitive to privacy.When mobile devices have strong mobility,different mobile devices frequently generate new tasks that require computation.Therefore,for privacy insensitive scenarios,the cache on edge servers can be effectively utilized to optimize task computation.However,considering the cache replacement of edge servers,it is also necessary to re use game theory to analyze the competitive relationship between edge servers.This article designs functions such as cache utility evaluation and cache replacement,and proposes an online cache resource allocation strategy based on potential games.In this strategy,using memory on edge servers to cache popular task data can improve task computing efficiency and reduce energy consumption on edge servers.Simulation experiments show that the proposed algorithm can effectively improve task completion rate,reduce energy consumption of edge servers,and improve edge server efficiency compared to other benchmark algorithms.