基于主观信任连通支配集的MANET安全分簇算法  

A secure clustering algorithm based on subjective trust connected dominating set for MANET

作　　者：廖俊[1,2] 蒋黎明[1] 张宏[1] 姜海涛[1] 

机构地区：[1]南京理工大学计算机科学与技术学院,南京210094 [2]中国药科大学信息管理与信息系统系,南京210009

出　　处：《高技术通讯》2012年第9期904-911,共8页Chinese High Technology Letters

基　　金：国家自然科学基金（60903027,61003210）,江苏省自然科学基金（SBK201022379）,高等学校博士学科点专项科研基金（20093219120024）和江苏省高校自然科学研究计划（10KJB520014）资助项目.

摘　　要：针对现有移动无线自组网（MANET）分簇算法普遍存在因不端节点干扰而导致网络性能急剧下降的问题，结合信任评价机制和连通支配集理论，提出了一种新的MANET安全分簇算法——最大主观信任连通支配集分簇算法（CAMCDS）。该算法首先采用直觉模糊相似度来评估节点的行为可信度，然后分布式实现可信连通支配集的构造，最终通过采用优先选举支配节点为簇首的策略实现安全分簇。分析和实验结果表明，该算法适于保障MANET层次路由的连通性和稳定性，且具有更低的通信开销和更高的安全性。To solve the problem of existing clustering algorithms for mobile ad hoc networks （MANET） that they lead to a sharp decline in network performance because they are vulnerable to interference from misbehaving nodes, a novel secure clustering algorithm for MANET, called the clustering algorithm based on subjective trust connected domina- ting set （CAMCDS） , was proposed in the study. It was created based on the trust evaluation mechanism and the connected dominating set theory. The algorithm estimates the credibility of node behavior with the intuitionistic fuzz- y similarity first, and then, distributedly selects the nodes with the maximum-subjective-trust undertaking domina- ting mission to construct connected dominating set, and finally, achieves the secure clustering by utilizing the strat- egy of electing the nodes in maximum-subjective-trust connected dominating set as cluster head. The analysis and experimental results show that the CAMCDS, has fewer communication overheads and higher safety, and more suit- ability for ensuring the connectivity and stability of hierarchical MANET.

关 键 词：移动自组网(MANET) 主观信任 直觉模糊集 极大独立集 成簇算法 

分 类 号：TP393.01[自动化与计算机技术—计算机应用技术]