基于行为的动态信任监控与更新机制研究

发布时间：2018-05-01 01:22

本文选题：开放网络环境 + 服务作业选择　；参考：《南京邮电大学》2014年硕士论文

【摘要】：随着网络技术的快速发展，网络环境呈现出开放、动态、异构的特点。为了达到高效、安全的作业效果，静态的信任机制已不能满足要求，因此动态信任机制的研究越发重要。首先本文运用用户相似群方法提出了基于动态信任的多层次服务选择模型。重点探讨了运用活跃度进行动态权重分配的综合信任计算方法，使得关于服务节点信任值的预测结果更加贴近真实值，从而为作业前的服务选择提供有效依据。然后本文分别从两个不同的角度提出了实时作业过程中动态信任更新机制的研究。一方面从用户节点角度出发，对影响用户信任值的四种因子：重要性等级因子、行为决策因子、时差奖惩因子、频次处罚因子进行深入分析，并基于这四种因子设计出一种细粒度的动态信任更新方法，建立支持行为上下文变化的用户级动态信任更新机制，达到实时监控和更新用户节点信任值的目的；另一方面，从服务节点角度出发，根据行为上下文信息对可靠属性、安全属性、可用属性和性能属性建立了认识，综合运用模糊聚类、信息熵理论和信任联盟技术得到科学的动态信任更新方法，建立了服务级动态信任更新机制，进而更加准确、有效的得到关于服务节点的实时信任值。最后，，本文设计了在线信任监控系统，通过具体的实例，验证了本文动态信任研究机制的科学性和有效性。
[Abstract]:With the rapid development of the network technology , the network environment presents the characteristics of open , dynamic and heterogeneous . In order to achieve the efficient and safe operation effect , the static trust mechanism cannot meet the requirements , so the research of the dynamic trust mechanism is more important .

In this paper , a multi - level service selection model based on dynamic trust is put forward by using the method of user similarity group . The comprehensive trust calculation method of dynamic weight distribution is discussed in this paper , so that the prediction result about the trust value of the service node is closer to the real value , so as to provide an effective basis for the service selection before the operation .

In this paper , the research of dynamic trust updating mechanism in real - time operation is presented from two different angles . On one hand , four factors influencing the trust value of users are analyzed in detail from the perspective of user node : importance level factor , behavior decision factor , time difference reward and punishment factor and frequency penalty factor , and a user - level dynamic trust updating mechanism supporting behavior context change is established based on these four factors , so as to achieve the purpose of monitoring and updating the trust value of user node in real time ;
On the other hand , from the point of view of service node , according to the behavior context information , we have established the knowledge of the reliable attribute , the security attribute , the available attribute and the performance attribute , and comprehensively applied the fuzzy clustering , the information entropy theory and the trust alliance technology to obtain the scientific dynamic trust updating method . The service level dynamic trust updating mechanism is established , and the real - time trust value about the service node is obtained more accurately and effectively .

Finally , this paper designs the online trust monitoring system , through concrete examples , validates the scientific nature and effectiveness of the dynamic trust research mechanism in this paper .

【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.06

【参考文献】