基于改进D-S证据理论的信任评估模型  

Trust evaluation model based on improved D-S evidence theory

作　　者：张琳[1,2,3] 刘婧文[1] 王汝传[1,2,3] 王海艳[1,2,3] 

机构地区：[1]南京邮电大学计算机学院,江苏南京210003 [2]江苏省无线传感网高技术研究重点实验室,江苏南京210003 [3]南京邮电大学宽带无线通信与传感网技术教育部重点实验室,江苏南京210003

出　　处：《通信学报》2013年第7期167-173,共7页Journal on Communications

基　　金：国家自然科学基金资助项目(61170065;61171053;61203217;61103195;61201163;61202354);江苏省自然科学基金资助项目(BK2011755;BK2011072;BK2012436);江苏省科技支撑计划(工业)基金资助项目(BE2011189;BE2012183;BE2012755);省属高校自然科学研究重大基金资助项目(12KJA520002);南京邮电大学科研基金资助项目(NY212063);江苏高校优势学科建设工程基金资助项目(yx002001)~~

摘　　要：针对已有的信任证据模型不能快速有效地处理分布式网络中存在的恶意攻击,且缺乏关于三元信任关系组的信任归一方法,提出了一种基于改进D-S证据理论的信任模型,在此基础上,提出了基于持续序列的基本可信度函数和基于评估函数的信任评估方法,使得模型能更快地抑制恶意节点,并且评估结果更贴近现实值。通过分析与仿真,验证了本模型具有抑制聚集信任攻击的有效性和健壮性,同时信任评估方法更具合理性和准确性。The existing trust evidence models cannot deal with malicious attacks in the distributed network quickly and effectively, and are lack of a trust normalization method that can measure trust relationship described by a triple set, so a trust model based on the improved D-S evidence theory was proposed. On this basis, basic trust value function based on continuous sequences and trust evaluation method based on evaluation function were also proposed to inhibit malicious nodes in a higher speed, and to enable the prediction results to be closer to the fact. Analysis and simulation show that this model has better effectiveness and robustness to deal with the aggregating trust attack, and trust evaluation method has better reasonableness and accuracy.

关 键 词：D-S证据理论 信任评估模型 基本概率分配函数 类概率函数 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]