Research On Computing Offloading Method For Edge Cloud

The popularity of mobile devices has allowed people to handle computing tasks from anywhere.However,as mobile applications become more powerful,the amount of computation required increases dramatically.Mobile devices are gradually unable to meet the needs of users due to limited local resources.In response to this problem,computing offloading techniques have been proposed.The mobile device sends the task with a large amount of computation to the cloud by performing computing offloading,thereby breaking the limitations of the hardware.However,because the cloud server is deployed in the network center,the service response time is longer.Therefore,the researchers propose to offload the computing task to the edge cloud server deployed at the edge of the network.This approach enhances the user experience by greatly reducing service response time.In addition,any user accessing the network has a risk of privacy leakage,and privacy issues when users perform computing offloading need to be taken seriously.Therefore,constructing a privacy protection computing offloading method for edge cloud is the current focus of research.In this thesis,the problems in the computing offloading strategy are deeply analyzed.By considering the real application scenario in the computing offloading process,the computing offloading strategy based on Stackelberg game and the computing offloading strategy based on privacy protection are proposed.The main research work of this thesis is summarized as follows:1.This thesis considers the computing offloading real-world application scenario for edge cloud,and proposes a computing offloading method based on Stackelberg game.Aiming at the problems in the existing research,this thesis introduces the classic game model in the field of economics,and transforms the process of computing offloading from the mobile device to the edge cloud server into the game process.In this method,the mobile device and the edge cloud server alternately propose their own strategies according to the strategy proposed by the other party,and finally reach Nash equilibriums through the two game algorithms proposed in this thesis,thereby obtaining the optimal strategies of both parties.The simulation results show that the method effectively improves the computing offloading efficiency of mobile devices and edge cloud servers,thus maximizing user benefits.2.Aiming at the privacy problem of two types of users in the process of computing offloading,a computing offloading method considering the privacy of users is proposed.This thesis introduces the differential privacy technology for this privacy problem and proposes two algorithms for different types of users.This method protects the privacy of mobile device and service providers by processing the data during the computing offloading process.Simultaneously,combined with the characteristics of the strategy selection of the computing offloading,this thesis introduces the online learning method,and the optimal strategy is obtained by analyzing and feedback the computing offloading behavior of mobile devices.The simulation results show that the method improves the computational offloading efficiency of mobile devices and improves the privacy protection level of users.In this thesis,some innovative results have been achieved by studying the hot issues in computing offloading from the aspects of strategy optimization and privacy protection.By analyzing the existing researches on computing offloading and considering the existing problems,a more comprehensive strategy selection method is proposed.At the same time,the simulation results show that the research results obtained in this thesis have a positive effect on the development of computing offloading.