基于DPI的数据内容安全检测算法优化与实现  

The Optimization and Implementation of Data Content Security Detection Algorithm based on DPI

作　　者：姚永锋[1,2] 梅瑞[1] 梅越 

机构地区：[1]北京大学软件与微电子学院信息安全系,北京102600 [2]浙江省公安消防总队作战指挥中心,浙江杭州310014 [3]空军司令部自动化工作站,北京100843

出　　处：《信息网络安全》2013年第7期2-6,共5页Netinfo Security

摘　　要：近年来,随着计算机技术的快速发展和互联网的迅猛普及,网络安全问题越来越受到人们的重视。文章首先介绍了DPI(Deep Packet Inspection,深度包检测)技术和几种常见的数据流检测算法,针对传统指纹检测算法的不足,提出了一种改进方案并在仿真平台上予以实现,然后研究了K取值不同时检测率的变化趋势,最后,完成了基于DPI技术的安全检测系统的设计,该系统由数据流监控引擎、Web控制台和安全中心三个模块构成,使得用户能够随时掌握系统的网络安全状况,效果显著。实验结果表明,优化后的指纹检测算法效率得到了大幅度提升并可以适应线上检测的要求,且K取值为16时,检测效果最好。Recently, with the rapid development of computer technology and Intemet, network security has got more and more attention. DPI technology and several commonly used Data stream Detection Algorithms were introduced and an improved scheme of traditional fingerprint detection algorithm was put forward because of its shortcomings. The improved scheme was implemented on the simulation platform and the trend of detection rate was studied when K has different values. In addition, a security detection system based on DPI technology was designed and the system consists of data stream monitoring engine, Web console and security centre, which enables users keep track of the network security situation and the effect is remarkable. The experiments prove that the optimized fingerprint detection algorithm can enhance the efficiency, adapt to the requirement of online detection and when the K value is 16, the detection effect is best.

关 键 词：DPI 指纹检测算法 安全检测系统 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]