计算机科学 ›› 2010, Vol. 37 ›› Issue (7): 110-114.

• 计算机网络与信息安全 • 上一篇    下一篇

基于分布式统计时间序列的网络流量分析

孟凡雪,刘衍珩,吴静,杨书奇   

  1. (吉林大学计算机科学与技术学院 长春130012);(吉林大学符号计算与知识工程教育部重点实验室 长春130012)
  • 出版日期:2018-12-01 发布日期:2018-12-01
  • 基金资助:
    本文受教育部高校博士点基金(20060183043)资助。

Analysis of Network Traffic Based on Distributed Statistical Time Series

MENG Fan-xue,LIU Yan-heng,WU Jing,YANG Shu-qi   

  • Online:2018-12-01 Published:2018-12-01

摘要: 研究网络数据在分布式存储下的相关性,有利于入侵检测整体的学习和指导优化数据的存储。重点研究了网络传输过程中各种类型数据的流量的这种相关性,提出了一种基于分布式统计(DS)的时间序列分析方法:根据网络协议间的关系将数据包分组,分析数量关系并给出报警阂值。仿真实验结果表明,该方法能较好地发现各种网络攻击。

关键词: 入侵检测,时间序列,分布式统计,网络流量

Abstract: Studying relationship of distributed storage data would be conductive to the overall intrusion detection learning,thus this relationship could not only be used for intrusion detection learning algorithm,but also supervise to optimite data's storage. According to the analysis focused on relationship of network traffic in the distributed storage, a method based on distributed statistical time series was proposed. According to the relationship of network protocol, this method could group data packet, thus analyzed quantity relationship and gave alarm threshold. The experiment results show that the method can be used to detect network attacks.

Key words: Intrusion detection, Time series, Distributed statistical, Network traffic

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!