Font Size: a A A

Research On Joint Radio And Computation Resource Management Of Mobile Edge Computing

Posted on:2021-01-18Degree:MasterType:Thesis
Country:ChinaCandidate:T YangFull Text:PDF
GTID:2428330614958252Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
With the rapid development of mobile communication technology and the wide spread of mobile intelligent devices,some new applications have emerged,such as augmented reality,face recognition,interactive games.These applications are delay sensitive and computationally intensive.For these applications,mobile devices with limited computation resource and endurance are difficult to provide users with ideal quality of service.The proposal of Mobile Edge Computing(MEC)provides a practical solution for the above problems.MEC builds a cloud computing environment at the edge of the network,and provides users with near-range computation offloading services by using radio and computation resources in the system.In MEC,radio and computation resources have an important impact on delay and energy consumption.However,the limitation of these two resources leads to fierce resource competition between users.Therefore,it is necessary to formulate reasonable task offloading strategy and radio and computation resource allocation strategy to achieve efficient use of resources.In this paper,the radio and computation resource management in single server and multi server scenarios of multi-user MEC system is studied.The specific work is as follows:For a single server MEC system,a utility based joint offloading decision and resource allocation algorithm is proposed,which is used to make the task offloading strategy,the power and computation resource allocation strategy for users in this scenario.Firstly,the problem is modeled as the problem of maximizing system utility under resource constraints,which is a mixed integer nonlinear programming problem.Then,the power and computation resource allocation strategies are formulated by using bisection method and KKT conditions respectively.Finally,on the basis of determining the resource allocation,the binary particle swarm optimization algorithm is used to develop the offloading strategy.Simulation results show that the average energy consumption of the proposed algorithm is 7.66% lower than that of the joint offloading decision and power optimization algorithm.Compared with the single server MEC system,the multi server MEC system also needs to consider the joint offloading decision between multi server and channel allocation.For multi server MEC system,a joint offloading decision and resource allocation algorithm based on improved differential evolution is proposed,which is used to make the offloading strategy,channel and computation resource allocation strategy for users in this scenario.The problem is modeled as a mixed integer nonlinear programming problem with resource and delay constraints.In order to solve the optimization problem,differential evolution algorithm is introduced,and the algorithm is improved according to the mixed integer characteristics of the problem,so as to develop offloading strategy and channel and computation resource allocation strategy.The simulation results show that the energy consumption of the proposed algorithm is 22.7% lower than that of the resource joint optimization algorithm based on independent offloading decision.
Keywords/Search Tags:mobile edge computing, resource management, task offloading, energy consumption, delay
PDF Full Text Request
Related items