Research On Resource Allocation And Service Migration Problem In Mobile Edge Computing

In recent years,the Mobile Edge Computing architecture has received extensive attention from academia and industry.By decentralizing part of the computing and storage resources in the remote cloud to the mobile edge computing server which is closer to the user,the architecture greatly reduces the transmission delay of the user requesting resources.To improve the user experience,the service provider preferentially allocates the computing task o?oaded by the user equipment to the edge computing server that is closer to the user equipment.However,due to the limited resources in the edge base station and the edge computing server,the computing tasks unloaded by the user equipment may not be fully processed internally.Therefore,it is especially important to develop appropriate resource allocation strategies and make full use of resources in the edge computing layer.At the same time,since the user equipment is constantly moving when using resources,when the distance between the user equipment and the edge computing server running the task increases,the communication delay also increases,and it is di cult for the service provider to guarantee the quality of service.The use of service migration technology to migrate services from a source edge computing server to an edge computing server which is closer to the user can e?ectively reduce transmission latency.However,the corresponding migration overhead occurs during the service migration process.Therefore,it is important to develop a suitable service migration strategy to balance migration loss and transmission loss.This paper discusses the resource allocation problem and service migration problem in densely deployed cellular networks under the mobile edge computing framework.In the formulation of resource allocation strategy,this paper considers the impact of the relationship between service providers,user equipment and edge computing servers on resource allocation strategies,and proposes a decentralized multi-service provider resource allocation algorithm.This algorithm ensures the quality of service provided by the service provider while maximizing the total revenue of the service provider at the edge computing layer.In the resource allocation problem,we comprehensively consider the allocation of computing and radio resources,and based on the matching theory,this problem is transformed into a matching problem between user equipment and edge base stations.The resource allocation algorithm proposed in this paper can dynamically adjust the resource allocation strategy according to the request of the user equipment and the distribution of resources in the edge computing layer.The experimental results show that the proposed algorithm outperforms the existing resource allocation algorithm in improving the total revenue of the edge computing layer.At the same time,this paper proposes a decentralized service migration algorithm for service migration in mobile edge computing.This algorithm dynamically adjusts the migration decision according to the location of the user equipment,related information of the o?oading task,the amount of resource idleness in the edge base station and the edge computing server.In this process,the service provider constructs an edge base station candidate set according to the location of the user equipment and the distribution of surrounding resources,and then selects an optimal edge base station and migration decision from the candidate set according to the established rules.The simulation results show that the service migration algorithm proposed in this paper can e?ectively reduce the total cost of the service provider to perform the tasks o?oaded by user equipment,and the e?ectiveness of our proposed algorithm is superior to the existing service migration algorithm in this problem.