Research On Task Offload And Joint Resource Allocation In Mobile Edge Computing

With the arrival of the 5G era,a large number of computationally intensive and delay sensitive applications that require low latency and low energy consumption have mushroomed.However,mobile devices have limited computing resources.Although traditional cloud computing platforms have abundant computing resources,data transmission latency is becoming increasingly difficult to meet user needs,and a large amount of data can cause network congestion,bringing huge pressure to the entire network.In order to improve the quality of user service and meet users’ demand for low latency and low energy consumption,Mobile edge computing(MEC)came into being.Compared to traditional cloud computing,MEC deploys servers at the network edge closer to mobile devices,providing computing and storage resources for mobile users,thereby improving their service quality and network usage experience.However,the amount of resources that MEC servers can provide is limited.With the increase in the number of mobile terminals,communication resources and MEC server computing resources are facing fierce competition.Therefore,there is an urgent need for effective task offloading strategies and resource allocation joint optimization methods to improve the efficiency of MEC system offloading and improve user network service experience.This article investigates the joint optimization problem of task offloading decisionmaking,communication resource allocation,and computing resource allocation in MEC network systems for two different application scenarios.The specific content is as follows:(1)In a multi user multi MEC server system,a joint optimization method for task offloading and resource allocation is proposed to reduce system latency and energy consumption for user sets with different latency sensitivities,while effectively utilizing limited spectrum and computing resources while meeting user latency requirements as much as possible.Firstly,this method adopts an averaging approach for task pre classification,and determines priority based on delay tolerance or interference to other users;Secondly,considering the peer effect,a externality matching algorithm is designed to match MEC server and subchannel resources for offloading users to reduce the interference between users on the same subchannel of different MEC servers;Then,Lagrange multiplier method under Karush Kuhn Tucker(KKT)condition is used to calculate resource allocation;Finally,combined with the above steps,a joint optimization algorithm is designed to set different unloading conditions based on user priority to obtain the optimal unloading decision for users.The simulation results show that the proposed method effectively reduces latency and energy consumption by 12% while achieving high satisfaction with user latency requirements.(2)In a cloud edge collaborative multi user single MEC server system,under the constraint of limited MEC server resources,a joint optimization method for computing task offloading decision and communication and computing resource allocation is proposed to improve offloading efficiency,reduce MEC system latency and energy consumption.This method fully considers the collaborative effect between remote cloud servers and edge servers,and divides the optimization problem into joint communication resource and computing resource allocation subproblems and offloading decision subproblems to solve.It is divided into four steps: first,use K-means clustering algorithm for task preprocessing;Secondly,allocate channel resources for offloading users based on delayed acceptance algorithms;Then,the Lagrange multiplier method is used to allocate computing resources for offloaded users;Finally,based on heuristic ideas,a joint optimization algorithm for task offloading and resource allocation was proposed to obtain the optimal offloading decision for each computing task.The simulation results show that the proposed method can improve unloading efficiency,reduce latency and energy consumption by 10.5%.This article proposes a joint optimization method for task offloading decisionmaking and resource allocation for two different application scenarios of MEC,with the goal of reducing system latency and energy consumption.The effectiveness of the proposed method is verified through simulation comparison.