Research On Optimizing Strategy Of Computing Offloading Energy Consumption Based On Lagrange Method

With the development of mobile internet technology and the popularization and widespread use of mobile intelligent terminals in people's daily lives,mobile terminals such as smart phones and tablets are becoming an important tool for learning,entertainment,social,news update and business communication.However,due to the limited resources(computing power,battery power,storage capacity)of mobile terminals,computationally intensive applications cannot run efficiently and permanently on mobile terminals.Driven by the Internet of Things and 5G communication,there has been an important shift in the field of mobile computing,from traditional centralized cloud computing to distributed mobile edge computing.Mobile Edge Computing Offloading(MECO)has become a cloud computing An important application scenario.Mobile edge computing is to integrate the computing,storage and other resources on the edge of the network into a unified user service platform.According to the principle of nearest service,the network edge node task request is timely and correspondingly processed.Computational migration is a key issue in edge computing due to limited edge node capabilities,resources,bandwidth,and energy.The core idea of the computational migration is that,according to the network state,the edge intelligent terminal migrates the computing task to other nearby mobile terminals to perform calculation and storage according to the carrying capacity,thereby realizing energy balance,reducing network delay and transmission overhead,and improving network QoE.Computational migration strategy has become one of the key issues in mobile edge computing services.The development of computing migration technology has introduced a new method to solve the problem of limited terminal resources.The work done in this paper on the issue of limited resources for mobile terminals is as follows:(1)The related research on computational migration is reviewed,and the characteristics and research priorities of computational migration at different stages of development are analyzed.This paper analyzes and compares several recent research results of edge computing,and summarizes the advantages and disadvantages of different computing migration resource partitioning strategies.(2)The system model of computational migration energy optimization is proposed,including system model,migration model and migration time constraint model.The system model of the thesis is a time-division-based MECO system based on Long Term Evolution(LTE).The system distinguishes user signals from different addresses according to time slots.In any time slot,the base station is based on Time Division Multiple Access(TDMA).A full/partial migration schedule for a subset of tasks for the user.The core idea is to optimize the energy consumption in the multi-user computing migration system,to dynamically divide the computing time into constraints,to balance the computing power of the system,and to circumvent the blind migration of a user in the multi-user migration system to save resources to ensure the total system.The calculation time is the lowest.(3)A Lagrange-based energy optimization model is proposed.According to the specific situation of the migration,the energy consumption of the terminal is divided into two parts: local computing energy consumption and calculation of migration energy consumption.The convexity proof of the energy consumption model is provided.The Lagrangian method is used to solve the model,and the Optimal Resource Partitioning Algorithm(ORPA)based on threshold is proposed.By comparing with the migration strategy proposed by YOU C S and the method of gradient descent method for solving Lagrangian,it is verified that the proposed model and the computational migration strategy ORPA can significantly reduce the energy consumption of mobile terminal task execution,effectively achieving the saving of mobile terminals.The purpose of energy consumption.And under the computational delay constraint,the execution time of the task can be significantly accelerated.On this basis,the paper also obtains the optimal number of users that the system can accommodate under different time slots,so as to set the time slots according to the needs of the actual application scenarios to meet the needs of different user populations.