面向VANET的声誉模型研究  

Research for Reputation Models in VANET

作　　者：陈辰[1] 韩伟力[1] 朱列[2] 王新[1] 

机构地区：[1]复旦大学计算机科学技术学院,上海201203 [2]上海交通职业技术学院,上海200431

出　　处：《小型微型计算机系统》2013年第2期233-237,共5页Journal of Chinese Computer Systems

基　　金：国家"八六三"高技术研究发展计划项目(2009AA01A348;2011AA100701)资助;国家自然科学基金项目(61171074)资助;复旦大学"九八五"三期学科建设项目资助

摘　　要：车载自组织网络是一个典型的开放式分布系统,其网络构成具有高度动态性和很强的临时性.这些特性导致车辆与合作节点之间缺乏相互了解与信任,并且可能面临各类攻击的威胁.针对上述问题,提出在VANET中建立面向应用的车辆节点声誉模型(简称VASRep),用于量化评估节点的声誉.VASRep充分融合车辆节点上丰富的数据类型,使得声誉数据更加准确地表现出节点的可信程度;VASRep依靠权威单元对节点进行评价的方法更好地适应高移动性的VANET环境.通过分析及仿真表明,VASRep较现有模型在性能表现、遏制更广泛类型的恶意节点攻击的能力等方面都有较大的提高.As a typical open distributed system, the Vehicular Ad hoc NETworks （ VANET ） are composed ephemerally due to their high mobility. Vehicle nodes in the VANET are facing with various malicious nodes and hard to choose trustworthy collaborative nodes in applications. To solve these problems, we design VANET application specified reputation models （ VASRep } to qualifiedly evaluate the reputation of the nodes. Through merging various types of information sources in VANET, VASRep can express the node＇s trustworthiness more accurately. According to the authority units＇ recommendation, VASRep is more suitable to VANET envi- ronments. Finally, the analysis and evaluation show that our model outperforms the existing solutions in terms of performance and re- sisting the attack of malicious nodes.

关 键 词：车载自组织网络 声誉 贝叶斯网络 权威单元 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]