一种新的大规模网络主动预警方法  

A New Approach for Large-scale Network Active Forewarning

作　　者：孙红杰[1] 方滨兴[1] 张宏莉[1] 

机构地区：[1]哈尔滨工业大学国家计算机信息内容安全重点实验室,哈尔滨150001

出　　处：《电信科学》2007年第1期74-78,共5页Telecommunications Science

基　　金：国家"863"高技术研究发展计划基金资助项目(No.2002AA142020);国家自然科学基金资助项目(No.60203021)

摘　　要：网络的全球化发展决定网络安全向整体性和动态性发展,互联网的分散管理机制使得监控管理域外的网络安全状态变得非常困难。本文基于网络透视技术,提出一种新的大规模网络主动预警方法,可以对无控制权的网络进行实时监控和主动预警。当网络发生异常行为时,能够准确检测异常来源,缩小异常定位范围,对于大规模网络异常行为的检测和预防具有重要现实意义。以中国教育网拓扑结构为背景的仿真实验表明,该方法有效,对网络安全具有重要的研究意义和实用价值。The network global development tendency decides that the network security evolution runs to all-side and dynamic, and it is difficult to monitor networks out of control because of Intemet decentralized control mechanism. In this paper, a new approach for large-scale network active forewarning is propesed based on network tomography, it can do the real-time monitor and make active forewarning for networks out of controL The sources of the abnormal trnflic can be detected and located in the beginning of their spreading. Experimental results based on CEBNET （China education and research network） topology indicate that the new approach is effective and has a further development potential for large-scale network active forewarning. It has important influence and practical value on the research of network security.

关 键 词：网络安全 大规模网络 主动预警 网络透视 网络模拟 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]