Research On Task Offloading And Caching Based On Edge Computing

With the popularity of smart devices,applications are widely used.With the continuous upgrade of applications,applications have higher and higher requirements for latency and bandwidth.Existing network systems are difficult to meet the requirements of further upgrading of applications,and edge computing effectively solves this problem by making full use of edge devices.In the actual scenario of edge computing,there are many problems that need to be solved,such as how to encourage devices to participate in the system,resource allocation and scheduling issues.This thesis mainly studies the task offloading and edge caching in edge computing.First,in the scenario of cooperative task offloading in edge computing,this thesis presents different problems for homogeneous mobile edge cloud and heterogeneous mobile edge cloud.Cooperative task offloading means that mobile edge clouds help each other in the form of offloading tasks,so as to achieve the purpose of improving system utility.In response to the above problems,this thesis designs a global auction mechanism for the problems in the homogeneous mobile edge cloud scenario,and a dual auction mechanism for the problems in the heterogeneous mobile edge cloud scenario.Finally,theoretical proofs verify the individual rationality and authenticity of the mechanism,and simulation experiments verify that the mechanism can enhance the effectiveness of the original system.Secondly,in the mobile edge cloud cache scenario,this article introduces the depth image-based rendering(DIBR)technology based on the characteristics of the multi-view video stream.The caching problem of mobile edge cloud is formalized as a 0-1 integer linear programming problem.In this thesis,the above problem is transformed into an equivalent set function optimization problem,and it is proved that this problem is a submodule maximization problem.Furthermore,this thesis proposes an improved 6)-cost benefit greedy algorithm.In addition,in order to solve the problem of insufficient resources of the mobile edge cloud,this thesis also proposes a simple incentive mechanism to help the mobile edge cloud.Finally,simulation experiments verify that the algorithm and mechanism can improve the effectiveness of the original system.