一种新的证据冲突分析方法  

A new method to analyze evidence conflict

作　　者：邓勇[1,2] 王栋[2] 李齐[2] 章雅娟[1] 

机构地区：[1]西南大学计算机与信息科学学院,重庆400715 [2]上海交通大学电子信息与电气工程学院,上海200240

出　　处：《控制理论与应用》2011年第6期839-844,共6页Control Theory & Applications

基　　金：国家自然科学基金资助项目(60874105);教育部新世纪优秀人才支持计划资助项目(NCET-08-0345);上海市青年科技启明星计划资助项目(09QA1402900);重庆市自然科学基金资助项目(CSCT;2010BA2003);上海交通大学国防科学基金资助项目(11GFF–17);西南大学博士基金资助项目(SWU110021)

摘　　要：当证据高度冲突时,使用DS组合规则经常会得到错误的结果,因此对DS证据理论的合成方法进行改进成为人们研究的重点.但是如何去度量或确定证据之间冲突的程度却常常被人忽视.在以往的研究中均使用冲突系数k作为表征证据之间冲突的量,但是研究结果表明,k不能很好的描述证据之间的冲突.对表征证据冲突的变量进行了研究,提出了一种新的表征证据之间冲突的变量-关联系数,基于偏熵和混合熵,定义了关联系数,该系数可以定量的表示证据之间的冲突.当关联系数接近1时证据之间冲突很小,当关联系数接近0时证据之间高度冲突.算例验证了所提出的基于关联系数的冲突表示方法的有效性.Dempster Shafer evidence theory is widely used in data fusion. However, incorrect results may be obtained by classical Dempster combination rule when collected evidence highly conflict with each other. Extensive work has been done to improve the DS combination rule, but how to measure the conflict among two pieces of evidence is an issue which has not been well considered. We define a new factor, the relative coefficient, for measuring the conflict between two pieces of evidence based on the partial entropy and the mixture entropy. When the relative coefficient is close to one, the conflict degree between two pieces of evidence is very low. On the other hand, when the relative coefficient is close to zero, the conflict degree between two pieces of evidence is very high. Some numerical examples are given to illustrate the efficiency of the proposed method.

关 键 词：D-S证据理论 组合规则 证据冲突 偏熵 关联系数 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置]