Research On Offloading Strategy Based On Mobile Edge Computing

In Mobile Edge Computing(MEC)technology,computing and storage resources are deployed at the edge of the network closer to mobile devices.Users offload computing tasks to the MEC server.MEC provides users with a high-bandwidth,low-latency IT service environment.As mobile communications enter the "5G" era,the number of users has increased sharply,and the types of tasks have become more diverse.How to ensure that users participate in mobile edge computing reasonably and how to allocate resources after offloading has become an urgent problem for MEC technology.In response to the above problems,this paper uses contract theory and Lyapunov optimization theory to solve the problem of resource allocation in mobile edge computing.The main content of this article can be summarized as follows:First,in the local MEC environment,the number of users is large,and the edge cloud server for computing and storage is limited.How to properly offload the tasks of mobile users to the edge server under the condition that the constraint conditions are met.By exploring the heterogeneity of user experience sensitivity,a ternary combination reduction model of CPU computing cycles-data volume-price is established,and a computing task allocation strategy based on contract theory is proposed to maximize the profit of cloud service providers while ensuring the non-negative benefits of mobile users.Second,the problem of computing resource allocation and power optimization of MEC server and Mobile Cloud Computing(MCC)server in the cloud-side collaborative system is studied.Based on the Lyapunov optimization theory,the power consumption of the cloud-side collaborative system is optimized.By establishing the queue dynamic equation and combining the objective function,a framework of Lyapunov drift + power consumption function is formed,and the problem is decomposed into three sub-problems,analyzed and solved one by one,and the optimal unloading strategy and resource allocation plan are obtained.Third,the resource allocation of MEC-based webcast platform is studied.The emergence of MEC technology has promoted the development of mobile web video live broadcast,and the video traffic has further increased sharply,and video service providers have obtained huge business opportunities.The webcast platform uses edge computing nodes to transcode and cache the user's video to reduce user delay and improve user experience.We design a payment plan for the video broadcast platform to the network operator(NO).The network operator(NO)leases its computing resources,such as computing capacity,spectrum and transmission power,etc.,and the video broadcast platform pays rent according to the amount of resources and quality of service(QoS).A multi-dimensional contract that considers the computing power,storage size,location,and bandwidth of the node is proposed to maximize the income of platform and at the same time make the NO actively participate.