期刊文献+

一种基于马氏距离的冲突证据组合方法 被引量:7

Combination Method of Conflicting Evidence based on Mahalanobis Distance
下载PDF
导出
摘要 D-S证据理论是高层数据融合中一种重要的处理不确定性问题的方法。但在使用D-S证据理论处理高冲突信息时,经常得出与直觉相悖的结论。证据理论不能有效处理高冲突证据的特点极大制约了其应用,当前研究主要对原始证据源进行修改。提出了一种基于马氏距离确权的组合方法,通过合理度量冲突大小,对原始证据源进行加权修正。经过数值分析,新的组合方法在应对敌对干扰等突发高冲突时性能得到明显改进。 D-S evidence theory is an important method for dealing with uncertainty in high-level data fusion.However,some counterintuitive conclusions are often acquired in handling highly-conflicting informations with D-S evidence theory.Due to this deficiency,the application of evidence theory is severely restricted,thus the current research aims at modifying the original source of evidence.This paper presents a combination method based on Mahalanobis distance,and the original source of evidence is corrected by properly measuring the conflict size.Numerical analysis indicates that,the new combination method enjoys significant performance improvement in dealing with high-conflict situations.
出处 《通信技术》 2015年第3期278-282,共5页 Communications Technology
基金 国家科技重大专项(No.2014ZX03006-003)~~
关键词 证据理论 证据冲突 马氏距离 evidence theory conflicting evidence Mahalanobis distance
  • 相关文献

参考文献7

二级参考文献80

  • 1叶清,吴晓平,宋业新.基于权重系数与冲突概率重新分配的证据合成方法[J].系统工程与电子技术,2006,28(7):1014-1016. 被引量:33
  • 2SHAFER G A. Mathematical Theory of Evidence[M]. Princeton, New Jersey: Princeton University Press, 1976.
  • 3YAGER R R. On the dempster sharer framework and new combination rules[J]. Information System, 1989, 41(2): 93 - 137.
  • 4SMARANDACHE F, DEZERT J. Proportional conflict redistribution rules for information fusion[M]//Advances and Applications of DSmT for Information Fusion (Collected works), Florentin Smarandache and Jean Dezert, Editors, Rehoboth: American Research Press, 2006,2:3 - 68.
  • 5MURPHY C K. Combining belief functions when evidence conflicts[J]. Decision Support Systems, 2000, 29(1): 1 - 9.
  • 6DENG Y, SHI W K, ZHU Z E et al. Combining belief functions based on distance of evidence[J]. Decision Support Systems, 2004, 38(3): 489 - 493.
  • 7BENJAMIN Q, MASSON M H, DENOEUX T. Refined classifier combination using belief functions[C]// Proceedings of the 11th International Conference on Information Fusion (FUSION '08). Piseataway: IEEE Press, 2008, 7:1 - 7.
  • 8JOUSSELME A L, GRENIER D, BOSSE E. A new distance between two bodies of evidence[J]. Information Fusion, 2001, 2(2): 91 - 101.
  • 9[1]Ronald R.Yager.On the dempster-shafer framework and new combination rules[J].Information Sciences,1987,41:93-137.
  • 10[2]G.Shafer.A mathematical theory of evidence[M].Princeton U.P.,Princeton,1976.

共引文献619

同被引文献75

  • 1吴根秀.信任函数组合与局部冲突处理[J].计算机工程与应用,2004,40(34):81-84. 被引量:17
  • 2郭华伟,施文康,刘清坤,邓勇.一种新的证据组合规则[J].上海交通大学学报,2006,40(11):1895-1900. 被引量:57
  • 3TAZID Ali, PALASH Dutta, HRISHIKESH Boru- ah. A New Combination Rule for Conflict Problem of Dempster-Shafer Evidence Theory [J]. International Journal of Energy, Information and Communications, 2012, 3(1) :35-40.
  • 4JOUSSELME A L, GRENIER D, BOSSE E. A New Distance Between two Bodies of Evidence [J]. Infor- mation Fusion, 2001(2):91-101.
  • 5HAN Deqiang , DEZERT Jean, YANG Yi. New Dis- tance Measures of Evidence Based on Belief Intervals [J]. Lecture Notes in Computer Science, 2014, 8764 (1): 432-441.
  • 6LIU W. Analyzing the degree of Conflict Among Be- lief Functions [J]. Artificial Intelligence, 2006, 170 (11) :909-924.
  • 7DEMPSTER A. Upper and Lower Probabilities In- duced by Muhivalued Mapping [J]. Annals of Math- ematical Statistics, 1967, 38 : 325-339.
  • 8YAGER R. On the Dempster Sharer Framework and new Combination Rules [J]. Information Sciences, 1987, 41(2) :93-137.
  • 9SMET P. The Combination of Evidence in the Trans- fer Belief Model [J]. IEEE Tran. On Pattern Analy- sis and Machine Intelligence, 1990, 12(5):447-458.
  • 10ZHANG Lianwen. Representation, Independence, and Combination of Evidence in the Dempster-Sha- fer theory[C]//Yager R R, Kacprzyk J, Fedrizzi M (Eds.). Advances in the Dempster-Shafer Theory of Evidence. New York : Wiley, 1994, 51-69.

引证文献7

二级引证文献24

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部