摘要
D-S证据组合规则可以在没有先验信息的情况下进行融合,这一优点使得D-S证据理论在多传感器融合系统中应用非常广泛。但BPA函数是建立在基本事件的幂集之上的,无法通过简单的统计方法来获得,为了解决这一问题,提出一种适应性更好的基于模糊数相似性的BPA生成方法。通过数据分类的例证说明该方法具有有效性和适用性。
In the framework of evidence theory,information fusion relies on the use of combination rule allowing the belief functions for the different propositions to be combined.A crucial role in evidence theory is played by Dempster ′s combination rule which has several interesting mathematical properties such as commutativity and associativity,and can be combined without prior information which make it widely used in many data fusion systems.However,how to construct basic probability assignment（BPA） is still an open issue.A new method to obtain BPA based on the similarity measure between fuzzy numbers is proposed.The efficiency and the applicability of the proposed method is illustrated by the illustration of Iris data classification.
出处
《现代电子技术》
2011年第15期5-7,共3页
Modern Electronics Technique
基金
中国航天科技集团公司航天科技创新基金(CASC200902)
西北工业大学种子基金(Z2010034)资助项目
关键词
多传感器融合
D-S证据理论
基本概率赋值函数
模糊数
multisensor data fusion
D-S evidence theory
basic probability assignment function
fuzzy numbers
