Research On Computation Offloading Models And Algorithms In Mobile Edge Computing

As an efficient decentralized key technology in the field of edge computing,the computation offloading can effectively supplement the insufficiency of limited computing capabilities of mobile terminals,and greatly reduce the time delay and energy consumption of the system,thus improving the Quality of Service(Qo S).At present,more and more computing is transplanted to the edge computing environment.The multiuser and multitask computation offloading is still facing the challenges of low real-time response and high energy consumption.This article investigates mobile edge computation offloading strategies,theories,and technologies from the multiuser and multitask computation offloading optimization models,algorithms,and methods.The main research work of this article is summarized as follows:(1)For the multiuser and multitask computation offloading problem in the stochastic environment of mobile edge computing(MEC),an energy-efficient multiuser and multitask computation offloading(EMMCO)optimization method is proposed.First,the EMMCO method takes into account the existence of dependencies between different tasks within an implementation,abstracts these dependencies as a directed acyclic graph(DAG),and models the computation offloading problem as a Markov decision process.Subsequently,the task embedding sequence in the DAG is fed to the RNN encoder-decoder neural network.With a combination of the attention mechanism,the long-term dependencies between different tasks are successfully captured by this scheme.Finally,the improved policy loss clip-based PPO2(IPLC-PPO2)algorithm is developed.The loss function in the IPLC-PPO2 algorithm is utilized as a preference for the training process,and the neural network parameters are continuously updated to select the optimal offloading scheduling decisions for tasks.The simulation results demonstrate that the EMMCO method can achieve lower latency,reduce energy consumption,and obtain a significant improvement in Qo S than the compared algorithms under different mobile edge network computing environments.(2)Aiming at the deficiency of high energy consumption and high resource overload probability in the multitask computation offloading process,an energy-resource-aware multitask computation offloading(ERMCO)optimization method is presented.First,the ERMCO method establishes a dynamic offloading optimization model for computation tasks.Then,by abstracting the mapping relationship between computation tasks and MEC servers into vectors,in order to calculate the vectors,the differential evolution algorithm is discretized,and an improved discreteness differential evolution(Discreteness-DE)algorithm is developed.Finally,the Discreteness-DE algorithm is employed to search the global optimal solution of the model in the solution space,so as to achieve the optimal offloading of computation tasks.The simulation results show that the ERMCO method is able to significantly reduce the system energy consumption,balance the resource utilization of the MEC server,and improve the Qo S of the system to a certain extent.(3)In view of the problems of high latency and high energy consumption in the computation offloading process of computation-intensive and latency-sensitive tasks in mobile edge computing,a priority-based computation offloading(PCO)method is suggested.First,the PCO method comprehensively considers the computing capability of the mobile terminal device,the latency and energy consumption of computing the task to judge whether the computation task needs to be offloaded.Then,the priority of the computation tasks that need to be offloaded and the priority of MEC servers are determined.Finally,according to the determined priority,the computation tasks with high priority are offloaded to the MEC servers with strong computing capability to complete the computation offloading.The simulation results show that compared with the traditional computation offloading algorithms,the proposed PCO method can effectively reduce the total latency and total energy consumption of the MEC system,thus improving the Qo S of the system.