上海交通大学学报(自然版)

• 自动化技术、计算机技术 • 上一篇    下一篇

一种基于数据流聚类的动态目标分群框架

龙真真1,2,张策2,王维平3,张正文4   

  1. (1. 国防科技大学 信息系统与管理学院, 长沙 410073;2. 空军装备研究院, 北京 100085; 3. 国防科技大学 研究生院, 长沙 410073;4. 中国科学院 数学与系统科学研究院, 系统科学研究所, 北京 100080)
  • 收稿日期:2009-07-06 修回日期:1900-01-01 出版日期:2010-07-28 发布日期:2010-07-28

A Dynamic Framework for TargetGrouping Based onClustering Data Streams

LONG Zhenzhen1,2,ZHANG Ce2,WANG Weiping3,ZHANG Zhengwen4   

  1. (1.School of Information System and Management, National University of Defense Technology,Changsha 410073, China; 2.Equipment Academy of Air Force,Beijing 100085, China;3.Graduate School, National University of Defense Technology, Changsha 410073, China;4.Institute of System Science, Academy of Mathematics and Systems Science,Chinese Academy of Sciences, Beijing 100080,China)
  • Received:2009-07-06 Revised:1900-01-01 Online:2010-07-28 Published:2010-07-28

摘要: 为了解决动态目标分群问题,提出了一种基于数据流聚类的动态目标分群框架.该框架分为在线和离线两部分.在线部分采用临时存储结构和金字塔时间框架保存侦察数据集的概要信息;离线部分采用CNM算法对时间框架的信息进行聚类,最终得到分群结果.实验表明,该框架具有灵活的精度和效率平衡性,能够较好地满足决策辅助系统处理实时信息的需要.

关键词: 动态目标分群, 数据流聚类

Abstract: In order to solve the dynamic targetgrouping problem, a framework based on clustering data streams was presented, which can be divided into two parts: online part and offline part. In online part, the concepts of a pyramidal time frame and a temporary storage structure are used; in offline part, CNM algorithm is used to cluster the suitable data. After the experiment, the results show that this framework has good equilibrium between accuracy and efficiency.

中图分类号: