Font Size: a A A

Research On Edge Computing Resource Allocation Algorithm Based On Game Theory

Posted on:2020-07-27Degree:MasterType:Thesis
Country:ChinaCandidate:H X WangFull Text:PDF
GTID:2370330620953257Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
With the continuous maturity of 5G communication,new storage systems,lightweight function libraries and other technologies,as well as the wide spread of terminal devices with smart chips,the era of Internet of Everything(IoE)has arrived,which makes people's demand for network data traffic and device computing power increasing.In order to meet these requirements,edge computing with local deployment,proximity to users,low latency and other characteristics has become the research focus of academia and industry.Devices deployed on the edge of the network provide rich computing resources and certain storage resources for the user terminals within its coverage.Due to the limitation of the coverage of edge devices and the resource capacity,the network still has the problem of load imbalance.Based on this,this paper studies the deployment of edge computing in the LAN composed of private small base stations,including:Firstly,a layered edge computing network architecture is proposed.First of all,based on "Terminal – Edge – Cloud",the Internet of things application oriented computing system integration development trend,and combined with the technical features of the different computing systems,and the hierarchical architecture of edge computer network was proposed,namely the terminal network(physical)and edge computing layer(virtual),each layer are analyzed in detail in such aspects as resource sharing,technical characteristics and demand.Secondly,an edge computing resource allocation algorithm based on game theory(ERAG)is proposed.For the long-term performance optimization of small base stations,the energy deficit queue is firstly established through the Lyapunov drift penalty technology,so as to limit the energy consumption in the long-term calculation process.Secondly,the system model is established according to the performance of the task,credit value,computing energy consumption and delay,as well as the congestion of the cable channel and other factors.The model aims to minimize the cost of small base stations to complete the task.Then,according to the characteristics of pure distributed deployment of small base stations,an edge computing resource allocation algorithm based on game theory is proposed,which realizes the Nash equilibrium of task unloading between small base stations through marginal cost and non-cooperative game,and obtains the optimal solution of optimization objective.Finally,the feasibility of the algorithm is verified by simulation.
Keywords/Search Tags:edge computing, architecture, game theory, Lyapunov, resource allocation
PDF Full Text Request
Related items