Research On Offloading Strategy Of Mobile Edge Computing Considering Privacy

The trend toward mobile edge computing(MEC)is gradually increasing to address the service needs of mobile devices,such as low latency and low energy consumption.As one of the key technologies in mobile edge computing,the function of computational task offloading is to migrate complex applications to run on the edge server-side.It is to solve the problem of insufficient resources and performance,such as storage and computation on the mobile side.Computational task offloading demonstrates high performance in achieving real-time data transfer and low energy consumption for computational tasks.However,due to the open nature of the network environment,computational task offloading can threaten user privacy and security.There is limited research in the literature on privacy protection during the offloading process in the MEC,so one of the critical issues that need to be addressed in current offloading techniques is how to use computational offloading techniques to reduce the computational latency and energy consumption of the task while at the same time protecting the privacy of the mobile device to some extent.After a comprehensive study and analysis of the state of the art of computational task offloading and privacy protection at home and abroad,the thesis studies and proposes a computing offloading strategy that takes into account the privacy security of mobile terminals for the two situations of multi-user single server and multi-user multi-server,which focuses on the efficient transmission in the offloading process,user privacy security,etc.,and the thesis mainly works as follows:First,in a multi-user single server scenario,a mobile edge computing offloading strategy that takes into account privacy tolerance is proposed for the different maximum values of privacy information leakage that mobile users can accept in an open environment.The policy introduces privacy tolerance to the traditional offloading decision model that considers latency and energy consumption.The pricing model is formulated for the limited computing resources at the edge server-side.At the same time,a benefit function matrix is established by combining the relevant parameters at the mobile and edge sides.The mathematical model developed is solved using knowledge of game theory,which reduces some of the computational and storage pressure on the server-side while protecting the mobile device’s privacy to a certain extent.Simulations show that the proposed privacy protection strategy protects the user’s privacy to a certain extent while reducing latency and energy consumption on the mobile side.Secondly,in a multi-user,multi-server scenario,an offloading policy based on user offloading frequency is proposed because different offloading frequencies of mobile users can make the identity be discovered.The policy quantifies the amount of privacy offloading from the user’s computational tasks,introduces a false task mechanism to protect the actual number of user offloads,and establishes a mathematical energy model that considers latency and privacy constraints.An improved particle swarm optimization algorithm is used to optimally solve the established mathematical model,keeping the cumulative privacy of different users at the edge server-side within a threshold,reducing the risk of privacy leakage and improving offloading efficiency.Simulation experiments show that,compared to an offloading strategy that does not consider the privacy scheme,this strategy in this paper leads to some increase in system energy consumption but is within the acceptable range for mobile users.At the same time this policy ensures real-time task transmission,weakens the accuracy of the attacker’s determination of the user’s identity,and provides some privacy protection.