分布式环境下散乱点云数据挖掘改进算法  

Improved data mining method for scattered point cloud data in distributed environment

作　　者：汪忠国[1] 吴敏[1] 

机构地区：[1]安徽工程大学机电学院,安徽芜湖241000

出　　处：《西安工程大学学报》2016年第5期633-638,共6页Journal of Xi’an Polytechnic University

摘　　要：在分布式云计算环境下,传统方法采用散乱点云数据模糊C均值聚类挖掘算法,在受到较强的数据特征干扰时,数据挖掘精度不高.为此,提出一种基于分布式云数据传输信道的散乱点云数据挖掘改进算法.通过构建散乱点云数据的存储结构,对数据传输信道进行多普勒扩展,降低数据挖掘过程中传输衰减损失;采用级联滤波算法进行数据干扰滤波,实现散乱点云数据的滤波提取,完成数据挖掘算法改进.仿真结果表明,采用改进的算法进行分布式环境下散乱点云数据挖掘,能有效提高散乱点云数据挖掘精度,频谱特征聚敛性能较好,抗干扰能力较强.In the distributed cloud computing environment,the traditional method uses the mining algorithm of fuzzy C means clustering of scattered point cloud data.When it is disturbed by the strong data features,the accuracy of data mining is not high.We proposed a mining algorithm based on scattered point cloud data mining algorithm of distributed cloud data transmission channel.The storage structure of the data was constructed to go on data transmission channel doppler extension,to reduce the transmission attenuation in the process of data mining.By using the method of cascaded filter to filter the data,to implement filtering extract of the data,to improve the algorithm.The simulation results show that the inproved algorithm can effectively improve the data accuracy.The convergent performance of spectrum characteristics is good,and the capability of anti-jamming is strong.It shows a good application value.

关 键 词：分布式环境 云计算 散乱点云数据 数据挖掘 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]