Research On The Capture And Allocation Of Computing Resources In MEC Network

With the development of communication technology and microprocessor technology,the popularization of smart devices has accelerated,causing the number of mobile devices to increase exponentially.On the one hand,the improvement of communication technology provides network support for emerging applications.On the other hand,the upgrading of microprocessor technology makes it possible for devices to handle high-energy,fastresponse applications.In addition,the Dynamic Voltage and Frequency Scaling(DVFS)technology embedded in the CPU enables the terminal device to have the function of local CPU self-regulation,which can further reduce energy consumption and extend terminal battery life by adjusting its own CPU Frequency according to the application load.At the same time,the emergence of emerging applications such as virtual reality has put forward higher requirements for the computing and battery life of mobile devices.However,existing mobile devices are still limited by physical size,battery technology,etc.,and limited computing resources and battery capacity.The emergence of Mobile Edge Computing(MEC)architecture provides a good solution to the problem of limited computing resources of mobile terminals—allowing the terminal to transmit complex applications to the edge cloud for processing and return the results-Application Offloading.However,the application uninstallation process brings additional transmission costs to the mobile terminal,including time delay and energy consumption.Device to Device(D2D)technology,as a key 5G communication technology,helps to achieve short-distance,low-latency,and low-energy data transmission.Application offloading combined with D2 D technology helps cache the transmission overhead in the offloading process,and further improves user experience.However,MEC has limited computing resources and cannot fully meet the computing needs of dense user scenarios.There are often a large number of idle terminals in the actual network,which can effectively supplement the shortage of edge cloud resources and improve the utilization of system resources.It is worthy of in-depth study.Therefore,how to make full use of idle terminals in the system,combined with D2 D technology and application offloading technology,to achieve effective computing resource capture(users use the computing power of nearby devices to perform tasks)and configuration research is very important.Below,this article will combine complex networks,application offloading,dynamic voltage adjustment,and D2 D technology to study resource capture-configuration algorithms under actual MEC networks to improve user experience and extend terminal battery life.The specific research content includes:1)Computer Resource Capture and Configuration Based on Network Motifs.Consider intensive MEC network abstraction as a complex network,compute-intensive applications from offloading to the free and work devices,analysis of different die bodies in the position in network,and put forward the computing resource capture algorithm based on network motifs,to capture the configuration of the system idle computing resources,raise the utilization ratio of network computing resources.2)An adaptive offloading configuration strategy integrating offloading rate,local CPU calculation speed and transmitting power.Considering the mobile edge computing system with limited computing resources,based on dynamic voltage and frequency adjustment technology,proposes a benefit-maximizing application offloading problem with corresponding delay,MEC computing resources and other network resource constraints,combined with variable replacement technology and sub-gradient descent Method,designed and proposed an Adaptive Offloading and Allocating Scheme(AOAS)that integrates offloading rate,local CPU calculation speed and transmission power.The simulation results show that when AOAS is configured,the system benefit can be increased by 22.25%compared with the greedy algorithm.