用于未知病毒检测的免疫识别模型和算法研究  

Research on Immune Recognition Model and Algorithm for Unknown Viruses Detection

作　　者：鲍欣龙[1] 马建辉[1] 罗文坚[1] 曹先彬[1] 王煦法[1] 

机构地区：[1]中国科学技术大学计算机科学技术系,合肥230027

出　　处：《计算机科学》2005年第1期74-76,94,共4页Computer Science

基　　金：国家自然科学基金(69971022);国家863基金(2002AA142130)

摘　　要：现有的反病毒技术难以识别与处理新的未知病毒。本文借鉴生物免疫系统识别未知病毒的机制,以非我识别机制为基础,进一步抽取免疫进化学习机制和阳性/阴性选择机制,提出了一种新的检测器和自我均自适应变化的免疫识别模型和算法。文中给出了算法的详细实现步骤,并针对几种实际病毒进行了检测实验。测试实验结果表明该算法能够检测到未知病毒,具有很好的应用前景。Lacking the ability of recognizing and processing the unknown viruses is one of the major limitations of common anti-virus technologies. Inspired by the vertebrate immune system,this paper puts forward and accomplished a novel immune recognition model and algorithm. Referencing the immune evolutionary learning mechanic and posi- tive/negative selection mechanic,this model and algorithm make the detectors and 'self' to change by self-adaptation. This paper gives out the detailed implementation of this algorithm. Experiments on several real viruses prove that this algorithm has a good ability on recognizing unknown viruses and wide prospective application fields.

关 键 词：未知病毒检测 免疫识别模型 算法 人工免疫 进化学习 计算机病毒 网络安全 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]