Research On Computing Offloading Method Based On Mobile Edge Computing

Mobile edge computing is characterized by geographic proximity to users to deliver high performance,high quality cloud computing services to users by deploying multiple mobile edge computing base stations to create a disruption-free,high speed,low latency network environment around users.Mobile edge computing is a reliable way for small and medium-sized computing devices,such as vehicles and cell phones,to be able to perform intensive computing that meets the needs of users.One of the key technologies of mobile edge computing is compute offload,which is a way to break the constraint of low computing power of mobile devices by handing over some or all of the computing tasks on mobile devices to the cloud computing environment to provide higher quality services to users.Most existing studies design a series of methods to accomplish desired goals based on the assumption of absolute rationality of users,but the actual results can deviate from the real situation because of the subjectivity of users.In this paper,we use a prospect theory framework to convert real-world decisions into mathematical models,and use Tversky and Prelec functions to simulate the profit risk avoidance psychology and loss risk preference psychology of people in the real world.In addition,the traditional artificial fish swarm algorithm is improved by using the current optimal fitness threshold qualification strategy,which enhances its ability to jump out of the local optimum and also ensures that the artificial fish swarm will not swim away from the optimal position.The simulation experimental results and the actual scenario experimental results show that the method proposed in this paper can better simulate the real scenario and can minimize the energy consumption while ensuring the time delay constraint.For the problem of locking users through device usage patterns,we propose a privacy-aware computational offloading method based on privacy entropy.By quantifying privacy as privacy entropy,the problem is modeled as maximizing privacy entropy while minimizing computational offloading resource consumption,and we propose to improve the Harris Hawk algorithm using Gauss-Cauchy operator to expand the search range of the algorithm and enhance the ability to jump out of the local optimum.Experiments demonstrate that the method can minimize resource consumption while ensuring sufficient confusion of user information,and can effectively solve the behavior pattern privacy leakage problem.