Research On Revenue Optimization Strategy Of Mobile Edge Computing Server

Mobile edge computing(MEC)is one of the key technologies of 5G.It places core functions such as computing,storage,and applications on the edge of the network near the terminal equipment or data source.It provides users with an IT service environment and cloud computing functions without going through the backhaul link and core network,which can meet users' needs in key aspects such as low latency,low power consumption,security and privacy protection.However,compared with cloud computing,the MEC server has limited computing resources and cannot meet the computing resource requirements of multiple users in heavy service load scenarios.In order to make full use of the computing resources of the MEC server,the industry has proposed a variety of resource allocation strategies aimed at rationally allocating the wireless resources and computing resources of the MEC server to maximize the user's benefits.However,there are few studies on how to use limited computing resources to maximize the benefits of MEC service providers.In fact,in order to make up for the cost of server deployment and maintenance,and achieve profitability,MEC service providers are very concerned about how to improve their own profits when computing resources are limited.In the heavy business load scenario,this paper studies the computing resource allocation strategy of the MEC server in two cases where application tasks are independent of each other and have sequential execution correlation,in order to maximize the benefits of MEC service providers.First,this paper considers a multi-user MEC system with a binary offloading.In this MEC system,each end user has a delay-sensitive task,and each task has a strict delay requirement.In order to ensure that the quality of service(Qo S)requirements of offloading users are met while increasing the revenue of MEC service providers,a strategy is proposed to improve the revenue of MEC servers by optimizing the execution order of computing tasks.The problem of maximizing MEC server revenue is modeled as an optimization problem with task execution order as the optimization variable,and then an algorithm based on branch and bound method is proposed to solve the task execution order.Simulation results show that,compared with the benchmark algorithm,the proposed algorithm can more effectively increase the average revenue of MEC service providers in heavy-load networks.Secondly,this paper considers the multi-user MEC system in which terminal application tasks have sequential execution correlation characteristics,and different users in this system have different degrees of preference for delay and money.In this scenario,how can MEC service providers maximize their revenue? In order to solve this problem,the MEC server's maximum revenue problem is modeled.According to the user's different degrees of delay and money preference and the network information that the MEC server has obtained,an algorithm based on ant colony algorithm to solve the optimal execution order of tasks is proposed.Then,obtain the optimal partition decision of the task and the MEC server execution order of the offload task.Simulation results show,the proposed algorithm can significantly improve the average revenue of MEC service providers.