Research On Trusted Collaboration Key Technologies For Cloud-Edge-End Environment

With the development of network technologies and applications,the data generated by terminal applications increases exponentially,which puts great pressure on centralized data processing and storage.Cloud-edge-end network architecture came into being.Cloud-edge-end network architecture extends computing services from cloud center to network edge through edge computing,and expands the scope of network computing.The architecture cooperates with the computing power of the cloud and the edge to provide intelligent services for mobile edge users nearby.It meets the key requirements of agile connection,real-time computing,data feedback,application intelligence,security and privacy protection,realtime access and optimization decision-making.The cloud-edge-end network ecological environment takes cloud services as the center,edge computing as the service extension,and edge devices work together.Since some key computing tasks rely on edge servers and edge devices,the control ability of the core network is weakened,and there are many security risks such as unauthorized access,sensitive data leakage,collaborative attack and so on.The trusted collaboration mechanism and technology for cloud-edge-end environment provide an effective way to solve the above problems.The research goal of this paper is to improve the security of cloud-edge-end services through an efficient and interactive cloud-edgeend trusted authentication mechanism,and provide trusted perception,evaluation and prediction methods for personalized cloud-edge-end collaborative interactive services in complex interactive environment.The main research work includes the following three aspects:(1)Aiming at the trust of mobile edge node interaction in cloud-edgeend network environment,a trusted cooperation mechanism of mobile edge nodes based on multi-source feedback is proposed.This mechanism comprehensively considers the rapidity,real-time,effectiveness and accuracy of mobile edge computing in the cloud-edge-end network environment,and can improve the cloud-edge-end collaborative computing service capability.Firstly,a trusted edge computing architecture based on cloud-edge-end network environment is proposed,which integrates the feedback factors from edge servers and other end users into the trusted collaboration evaluation of mobile edge devices,making the proposed trusted collaboration mechanism more reliable.Secondly,a lightweight trusted collaboration mechanism based on objective information entropy theory is proposed,which has higher computing efficiency and lower overhead,and is more suitable for large-scale cloudedge-end network environment.Then,a reliable and adaptive overall trusted collaborative aggregation algorithm for mobile terminal equipment based on objective information entropy theory is proposed,which can overcome the subjectivity of the traditional information fusion algorithm based on information entropy.Finally,the experimental results verify the advantages of the trusted collaboration mechanism in computing overhead and security.(2)Aiming at the monitoring efficiency of large-scale service behavior in cloud-edge-end network environment,a fast parallel trusted computer system for collaborative cloud services based on service perception is proposed.This mechanism can realize fast,parallel and distributed trusted collaborative computing,and can be widely used in large-scale collaborative services in cloud-edge-end network environment.Firstly,based on the distributed monitoring agent,a distributed modular sensing architecture for the service behavior of large-scale virtual machine cluster is proposed.Through distributed and modular design,the architecture can quickly perceive the service behavior of virtual machine cluster in the cloud-edge-end environment with huge network entities.Secondly,based on the large-scale,real-time,dynamic and multidimensional behavior data perceived by distributed agents,an adaptive,lightweight and parallel trusted computing scheme is proposed.According to the inherent time attenuation characteristics of trust relationship,a new algorithm combining time window mechanism and time attenuation function is used to calculate the reliability of virtual machine,which can effectively meet the accuracy requirements of reliability calculation.At the same time,due to the adoption of distributed parallel computer system,the speed of trust calculation is greatly accelerated,which makes the trust calculation scheme very suitable for large-scale cloud-edge-end network environment.The prototype system and experiments verify the performance advantages of this mechanism compared with the previous schemes.(3)Aiming at the accuracy of service resource matching in cloudedge-end network environment,a trusted perception and fast resource matching mechanism for personalized collaborative cloud service is proposed.The mechanism adopts a trusted aware cloud-edge-end service agent architecture,which can not only provide the calculation of the reliability between collaborators,but also carry out personalized resource matching.At the same time,in order to improve the efficiency of personalized resource matching,a matching method based on cloud resource clustering is proposed.By analyzing the multi-dimensional service quality index,this method clusters the heterogeneous resources,reduces the search space of matching tasks,and then improves the efficiency.Secondly,based on multi-dimensional,real-time and dynamic monitoring data,an efficient,fast and adaptive service resource security and trusted collaborative computing algorithm is proposed,which effectively meets the needs of users for reliability and fast service of collaborative computing.Based on the above scheme,the prototype system is realized.The experimental results show that the proposed trusted perception and fast resource matching mechanism for personalized collaborative cloud service is superior to other existing solutions in terms of computational efficiency,accuracy and service satisfaction.