Research On Intelligent Offloading Approach And Trust Enhancement Method For Edge Computing Of Iov

In the edge computing of the Internet of Vehicles(IOV),due to the resources of the devices in edge layer are limited,and the types of processing tasks are different,it is necessary to use intelligent decisionmaking method to make full use of the limited computing resources and improve the timeliness of offloading decision.In addition,trust is a necessary condition for building a collaborative environment.Wireless,open and distributed environment makes edge computing challenging in trust and security.It is very important to establish a reliable and efficient trust management mechanism.In this thesis,a container hierarchical collaboration based integrated concurrent pattern offloading approach is proposed firstly.By using the method of subtask offloading,the container-layering cooperative offloading scheme is improved,and an edge computing offloading model based on container hierarchical collaboration is built.In order to minimize the total offloading delay,an Integrated Concurrent Pattern Artificial Bee Colony(ICP-ABC)algorithm based offloading approach is proposed.The experimental and simulation results show that the offloading model alleviates the problem of limited selection of available target offloading nodes caused by the difference of processing capacity of edge devices and the heterogeneity of services provided,improves the resource utilization of edge devices and improves the bearing capacity of the system when the offloading demand pressure is large.The ICP-ABC algorithm is used to make the offloading decision,improve the decision speed,and find the decision result with lower total offloading delay,and has better stability.There is also an edge offloading trust enhancement method based on blockchain proposed in this thesis,which is realized by the trust enhancement mechanism based on IFCM(Improved Fuzzy-C Means)clustering algorithm and TLGPCA(Trust Level and Graph Partition based RPCA)lightweight consensus mechanism.In the trust enhancement mechanism,the shared resources and historical behavior of one node is comprehensively considered to renew its reputation value.IFCM clustering algorithm is used which includes initial membership mechanism customizing,category separation introducing and cluster revising to divide the trust level and grant permissions to devices in IOV,so as to improve the ability to identify malicious nodes.TLGPCA is proposed to complete the consensus and its graph partition mechanism divides the whole region with the goal of evenly distributing all types of nodes and minimizing the number of cross partition transactions.Then,the list of trusted edge computing nodes in the whole region at this stage is updated and saved through intra partition transaction consensus,inter partition transaction consensus and inter partition block consensus.The experimental and simulation results show that the above method is consistent with the theoretical analysis,and numerical results demonstrate that this method can improve transaction throughput,reduce average communication overhead,reduce transaction response time,and maintain the robustness of consensus.