基于模糊数学和主成分分析的数据融合算法研究

Research on Data Fusion Algorithm Based on Fuzzy Mathematics and Principal Component Analysis

本文利用多元统计方法中的主成分分析法计算测量值与主成分的相关系数,得出各传感器的综合支持程度,剔除综合支持程度较低的数据,取得有效数据。对有效数据,利用多传感器数据模糊相关特性,提出了一种基于模糊贴近度的数据融合方法。该数据融合方法计算过程相对固定,计算量小,便于计算机实时实现。

In this paper,we use the principal component analysis in the multivariate statistical method,count the correlation coefficient between the measured value and the principal component,the comprehensive support degree is obtained,the lower comprehensive support level data are ex-cluded.We acquired valid data.For valid data,using fuzzy correlation characteristics of mul-ti-sensor data,a method of data fusion based on fuzzy closeness is proposed.The calculation process of the data fusion method is relatively fixed and the calculation amount is small,which is easy to be realized by computer in real time.