Research And Application Of MEC Computing Offload Strategy

The Industrial Internet of Things is on the rise.At present,there are more than 600 million Internet of Things(IoT)devices.Industrial IoT devices also account for a large proportion of them.The IoT devices will transfer the collected data to the server.The data type and the large number of data characteristics are notable.However,the computing power of industrial equipment is limited and cannot handle the needs of existing application scenarios.The introduction of cloud computing solves the problem of insufficient computing power of industrial equipment,but cloud computing brings a series of problems such as data transmission costs,cloud storage costs,Internet access management,and security.Industrial equipment transmits data to cloud servers,which consumes transmission delays.And it is difficult to meet the requirements of low latency.Mobile Edge Computing(MEC)proposes to sink the computing power of the cloud server to the IoT device.Firstly,it can ensure low latency,and secondly,when the device requests task processing,it can quickly respond to the device.This article focuses on the research and application of computing offload strategy in MEC.Based on the basic Particle Swarm Optimization,this paper proposes Particle Swarm Algorithm Optimization(PSAO).The industrial Internet of Things scenario is modeled as a multi-user and multi-MEC optimization delay problem.Under constrained conditions,the total cost of reaching the entire system is minimal.The Industrial Internet of Things management and control system is built through the MEC platform.The PSAO offload strategy serves as the core algorithm of the scheduling module of the management and control system.The optimal allocation of tasks is obtained through the scheduling module.The client publishes data to different MEC servers based on the MQTT protocol.This article first describes the domestic and international status of mobile edge computing and computing offload strategy,introduces the MEC reference structure and key technologies,and analyzes the needs of industrial application scenarios.Secondly,I deeply researched the calculation offload strategy in MEC,modeled the actual scene of Industrial Internet of Things as delay model,energy consumption model and calculation model,and found the offload vector through the particle swarm offloading process,so that the total cost of the current system was minimized.Finally,based on the MEC platform,an industrial Internet of Things management and control system was built,which realized data analysis,data storage,and reverse control of equipment in the Industrial Internet of Things.The PSAO offload strategy was applied to the scheduling module to achieve a reasonable distribution of tasks.The MEC simulation platform built by OpenStack was used to test the performance of industrial scenes.The edge computing and cloud computing were compared in terms of delay,packet loss rate,and bandwidth,and the edge computing platform's superior data processing performance was verified.The PSAO offload strategy is compared with the local offload strategy,MEC benchmark offload strategy,and Artificial Fish-Swarm offload strategy.The results show that the PSAO offload strategy can effectively reduce the total system cost in the industrial Internet of Things environment.