一种基于度量层信息的基本信任分配构造方法  

A Basic Belief Assignment Construction Method Based on the Information of Measurement Level

作　　者：贾宇平[1] 杨威[1] 付耀文[1] 庄钊文[1] 

机构地区：[1]国防科学技术大学电子科学与工程学院,长沙410073

出　　处：《电子与信息学报》2009年第6期1345-1349,共5页Journal of Electronics & Information Technology

摘　　要：在决策层融合目标识别中,Dempster组合规则是一种常用的融合算子,它的有效应用取决于相应基本信任分配的合理建立。该文针对决策层融合目标识别问题中子源传感器输出的度量层信息,分析了基本信任分配的构造准则,并提出了一种基于参考向量与度量层信息之间相似度的基本信任分配构造方法。在仿真数据与空中雷达目标实测数据上的实验结果表明,该方法构造的基本信任分配能够有效地通过Dempster组合规则进行融合。Dempster rule of combination is a useful fusion operator for decision level recognition fusion. The effective application of this rule depends on the rational construction of corresponding basic belief assignment. Considering the measurement information from each sensor in decision level recognition fusion, this paper suggests a group of principles for constructing basic belief assignment, and then presents a strategy of basic belief assignment construction based on the similarity degree between referential vector and the measurement information. The experiments on artificial data and real data of radar aerial target demonstrate the basic belief assignments constructed by the presented method can be fused by Dempster rule of combination effectively.

关 键 词：目标识别 基本信任分配 DEMPSTER组合规则 度量层信息 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]