摘要
为了有效融合高度冲突的证据,提高证据高度冲突时融合结果的可靠性和合理性,给出了一种新的加权证据组合方法。该方法根据证据之间的关联特性,通过计算证据体间的相对距离获得系统中各传感器证据的权重。首先引入一个证据体间的距离函数,根据多证据体的距离,得到系统中各传感器证据体之间的相似度和各证据的支持度,依据支持度的大小进一步可得到各证据的可信度,以此作为各证据的权重,对证据的基本概率指派进行加权平均,最后利用Dempster组合规则融合加权平均后的证据。与其他方法相比,该方法在系统存在伪证据(干扰)时能有效迅速地识别出目标。
In order to efficiently combine high conflict evidence and improve the reliability and rationality of the combining results when evidences are in high conflict, a new method of weighted evidence combination rule is proposed. In adopting this proposed approach, according to the associated characteristic between evidences, the weight of evidence in sensor system is acquired by calculating the relative distance in body of evidence. First, a distance function between the bodies of evidences is introduced. According to the bodies of evidences, the similar degree between the bodies of evidences from different sensors and the support degree of each evidence are gained. Then the reliability of each evidence can be got from the support degree, with this as the weight of each evidence in weight - averaging the basic probability assignment. Finally the fusion of the weight - averaged evidences is performed with the Dempster combination rule. Compared with the existing methods, the proposed approach is more efficient and rapid in recognizing the target when false evidence is in existence.
出处
《空军工程大学学报（自然科学版）》
CSCD
北大核心
2008年第6期57-61,共5页
Journal of Air Force Engineering University(Natural Science Edition)
基金
国家自然科学基金资助项目(60773209)
关键词
证据理论
组合规则
冲突证据
目标识别
evidence theory
combination rule
conflict evidence
target recognition
