分布式多空间数据库信息快速融合方法仿真

摘要

This research focuses on the method for fast fusion of information of distributed multi-spatial database based on multi-attribute decision.Fuzzy theory was research used to extract information feature of each space in database and each information feature was defined as an input quantity of fuzzy proposition to obtain judging support degree of the fuzzy proposition.Then degree of information optimization and comprehensive confidence degree of feature recognition of each information source were acquired.The distance among each space information in system via Euclidean distance was measured to obtain entropy of decision significance representing information sequence.In addition,local fusion weight of each information attribute was confirmed and global fusion weight of each spatial information resource was obtained on that basis.Experimental simulation proves that the method has high fusion efficiency and improves the use ration of information resource effectively.%在分布式多空间数据库中最重要的部分就是对各个空间数据的共享和集成,因此需要进行数据库信息快速融合方法的研究.但是采用当前方法进行信息快速融合时,无法给出信息环境变化的动态特性,存在信息融合效率低的问题.为此,提出一种基于多属性决策的分布式多空间数据库信息快速融合方法.方法利用模糊理论提取数据库中各个空间的信息特征,将每个信息特征定义为模糊命题的一个输入量,得到对模糊命题的判断支持度,获取各信源的信息优化度与特征识别综合置信度,通过欧式距离衡量系统中各个空间信息之间的距离,得到表征信息序列的决策重要性的熵值,确定各信息属性的局部融合权重,在此基础上得到各个空间信息源的全局融合权重,以此为依据完成对分布式多空间数据库信息快速融合.实验结果表明,所提方法融合效率高,有效的提升了对分布式多空间数据库信息资源的利用率.