一种改进的动态信任信息聚合算法  

An Improved Aggregation Algorithm of Dynamic Trust Information

作　　者：王春红[1] 任姚鹏[1] 姚喜妍[2] 

机构地区：[1]运城学院计算机科学与技术系,山西运城044000 [2]运城学院应用数学系,山西运城044000

出　　处：《计算机技术与发展》2016年第3期117-120,共4页Computer Technology and Development

基　　金：国家自然科学基金资助项目(11241005)

摘　　要：随着分布式应用系统的发展,系统形态正从相对静态的形式向开放的、动态的服务模式转变。在这动态环境中,基于证书的静态信任机制就不再适应,就出现了动态信息机制管理问题。其中,核心问题之一是动态反馈信息的聚合。而现有的动态信任聚合算法基本都采用基于信任链或广播方式,这样往往导致了系统运算收敛慢、可扩展性差等问题。文中提出一种改进的动态信任信息聚合算法。首先,利用一种反馈信任值聚合方法获得反馈信任值;接着,在此基础上,基于蚁群算法通过多次循环寻找出多条较优的独立信任路径,并获得最终的反馈信任值。实验结果表明,该算法效率较高,可扩展性较好,可以获得较合理的反馈信任值,而且可在一定程度上有效防止联合欺诈行为。With the development of the distributed application system, the system form is changing from the relatively static form to the open and dynamic server mode. Among the dynamic environment,the trust mechanism based on the certificate is no longer adapted. So the problem of dynamic mechanism of information management issues. Always based on the trust chain or the broadcast way, current dynamic trust aggregation algorithm leads system operation convergence slowly and scalability badly. In this paper, an improved aggregation algorithm of dynamic trust information is proposed, which uses a aggregation method of feedback trust values to calculate the feedback trust value and adopts a method of searching trust path based on ant colony algorithm which can choose many better independence paths by a few circles,and then can obtain the final feedback trust value. The experimental results show that this algorithm has high efficiency and good scalability, and can get a reasonable feedback trust value, and prevent unites cheat behavior in a certain extent.

关 键 词：动态信任 反馈信任 蚁群算法 信任路径 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]