【作者】 周学广； 张焕国；

【Author】 Zhou Xueguang 1,2,Zhang Huanguo 2 1.College of Electronic Engineering,Naval University of Engineering,Wuhan 430072,P.R.China 2.School of Computer,Wuhan University,Wuhan 430033,P.R.China

【机构】 海军工程大学电子工程学院； 武汉大学计算机学院；

【摘要】 网络内容安全研究目前受到网络突破技术的挑战。中文网络突破技术之一是在关键词中恶意夹杂字符,和/或用中文同音字/繁体字代替关键词中的汉字,造成网络上极少数恶意网页成功地逃避了网络过滤和入侵检测,而常见的字符串匹配算法无法检测出这类攻击。文章给出了柔性中文字符串匹配的形式化定义,比较了经典字符串匹配算法,选择用前缀搜索方法实现了单模式串和多模式串的柔性匹配算法,完成了恶意夹杂字符的中文字符串匹配,部分地解决了其他传统字符串匹配算法无法解决的中文网络突破这一难题。更多还原

【Abstract】 Presently the research on the security of network contents has faced the challenge of network breakthrough techniques.A popular Chinese network breakthrough technique is to jam malicious characters in ,and/or replace the Chinese characters in key words with homonyms or homophones,or complicated Chinese characters.With this method some malicious web pages make themselves escape from the network filtering and intrusion detection.While most existing pattern matching algorithms in Chinese strings cannot detect this kind of attacks.In this paper we give a formulized definition of flexible pattern matching in Chinese strings,and compare the classical algorithms for pattern matching in strings,and choose the prefix searching strategy to realize the flexible algorithms for pattern matching in single-pattern string and multi-pattern string.Our algorithm can match the Chinese strings jammed with malicious characters,and partially resolve the Chinese network breakthrough problem that other classical matching algorithms in string cannot solve.更多还原