基于信息融合的网络流量监控研究

Study of network traffic monitoring based on information fusion

鉴于现有的网络流量监控系统向用户展现过多的网络信息往往无助于安全事件和网络故障的发现和解决,提出一种基于信息融合的网络流量监控框架.基于该框架可有效评估由网络负载状态、安全状态和故障状况构成的网络综合态势,给出了一种信息融合的网络流量监控系统设计,重点讨论了该系统中的数据预处理、关联分析和态势评估三个关键问题.该系统把海量数据融合成信息和知识,直观反映了网络的宏观态势、威胁,能精确定位网络故障.

Considering that the existing traffic monitoring system show users excessive network information, which is often helpless to find the security events and network failures and to solve them. This paper presents a information infusion-based framework for traffic monitoring, based on which, the network’s comprehensive situation composed of network load, safety sate and failure status can be effectively assessed. It also gives the design of traffic monitoring system with data infusion, and stresses three key problems in this system, including data preprocessing, correlation analysis and situation assessment. This proposed system can mix together massive information into useful information and knowledge, reflects intuitively the network’s macro situation and threaten, and locate its failures accurately.