Research On Computation Offloading Scheme In Mobile Edge Computing

With the rapid development of technologies such as the Internet of Things,big data,and 5G,cloud computing is unable to meet the application service requirements of large flow and large connections.Mobile edge computing(MEC)provides an effective solution for business localization by transmitting data from mobile terminals to the network edge through computation offloading technology.Designing effective computation offloading schemes for different application requirements and improving network quality of service are important research issues for MEC computation offloading.However,the computing power,storage space,battery energy,and other resources of MEC nodes are limited,making it difficult to ensure the long-term online of nodes while satisfying real-time applications.In this regard,we propose a MEC computation offloading scheme for joint optimization of delay and energy consumption is proposed to solve the problem of computation offloading under low delay and low energy consumption requirements and improve the network quality of service.Additionally,we consider the influence of factors such as the mobility and selfish rational behavior of MEC nodes on computation offloading.To illustrate the problem,we study the computation offloading of high-definition video streams for intelligent ambulances during the process of performing first aid,specifically during the scene of’getting on the car and admission’.We propose a MEC computation offloading scheme for intelligent ambulances to realize the transmission of large flow,high reliability,and low delay of emergency applications in intelligent ambulances.The main research work of this paper is as follows:1)To solve the problem of joint optimization of MEC computation offloading delay and energy consumption,we propose a MEC computation offloading scheme for joint optimization of delay and energy consumption,taking into account the influence of edge server load on computation offloading delay and energy consumption.To minimize task execution delay,terminal energy consumption,and edge server load rate standard deviation,we construct an MEC computation offloading cost optimization model.Secondly,we design a multi-mutation differential evolution(MDE)algorithm with multi-mutation operator and updating mutation operator with iterative correlation probability to solve the optimal offloading decision and achieve the optimal computation offloading cost.To verify the effectiveness of the MDE algorithm,we construct three different scale experimental networks based on Autonomous Systems by the Skitter public dataset.The simulation results demonstrate that compared with the random computation offloading scheme,energy optimization computation offloading scheme,and multi-objective greedy computation offloading scheme,the MDE algorithm improves the execution success rate,offloading success rate,and server load balancing by 13.23%,12.96%,and 29.37%,respectively.This realization of efficient,low-cost,and load-balanced computation offloading in MEC is significant.2)To address the problem of computation offloading of high-definition video streams for intelligent ambulances,we propose a MEC computation offloading scheme for intelligent ambulance(COS-IA).Firstly,we construct a comprehensive performance evaluation model of MEC nodes by integrating trust,stability,and available load rate.Secondly,we identify the available MEC nodes based on the driving path and direction of the intelligent ambulance.We construct a computation offloading priority evaluation model of MEC nodes and integrate comprehensive performance and priority evaluation to construct the candidate collaborative service set of intelligent ambulance computation offloading.We design the fitness function with delay,load,and link stability and propose an improved ant colony multipath computing offloading collaborative service set optimization algorithm to find TOP-N disjoint computation offloading collaborative paths in the candidate collaborative service set.Moreover,a dynamic adaptive balanced multi-path computation offloading task scheduling scheme is designed to realize the adaptation of large flow computation offloading of intelligent ambulance to multi-path collaborative path load and improve the stability of computation offloading.To evaluate the performance of our model,we simulate and analyze it based on a public dataset of a city road network.The simulation results show that the average bandwidth consumption of the offloading cooperative node of COS-IA is 400.71 Mbps.Compared with the stochastic routing algorithm,local area network stochastic routing algorithm,and local area network on-demand multipath routing algorithm,the success rate of offloading of COS-IA is increased by 47.57%,37.51%,and8.87%,respectively.Additionally,the stability of offloading is increased by 12.88%,38.54%,and3.44%.This meets the transmission requirements of large flow,high reliability,and low delay in emergency applications of intelligent ambulances.