基于多无线传感器的改进数据融合算法  

Improved Data Fusion Algorithm Based on Multiple Wireless Sensors

作　　者：杨金鑫 王志明[1] YANG Jin-xin;WANG Zhi-ming(School of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing 210000,China)

机构地区：[1]南京理工大学机械工程学院,江苏南京210000

出　　处：《仪表技术与传感器》2020年第8期118-121,126,共5页Instrument Technique and Sensor

基　　金：国家重点研发计划资助项目(2018YFD0701005,2018YFD0700800)。

摘　　要：为了提高采集数据的准确性和避免查表,对自适应加权融合算法进行了改进。根据正态分布中中值和平均值的特性,提出预处理剔除异常值的新方法,进而提高最终融合结果的准确性。利用该算法对某温室的待处理温度数据进行了计算,结果表明:与传统的格拉布斯法剔除异常值等算法相比,该改进算法能够使融合结果的标准差更小,融合结果更能代表整体的分布,同时无需查表。In order to improve the accuracy of the collected data and avoid table lookup,the adaptive weighted fusion algorithm was improved.According to the characteristics of the median and the mean in the normal distribution,a new method for preprocessing the outliers is proposed,which improves the accuracy of the final fusion results.The algorithm is used to calculate the temperature data of a greenhouse.The results show that compared with the traditional Grubbs method,the improved algorithm can make the standard deviation of the fusion result smaller,and the fusion result is more representative.The overall distribution,while no need to look up the table.

关 键 词：无线传感器 温度采集 数据融合 自适应加权算法 粗大误差 正态分布 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]