摘要
为提高未知指纹对应操作系统的识别速率和准确率,文中提出了基于奇异值分解和使用有向无环图分类的操作系统类型识别新方法。首先对操作系统原始指纹生成的矩阵进行奇异值分解提取奇异值特征;然后使用有向无环图生成多类分类器对未知指纹的奇异值特征进行分类;最后在Nmap指纹库上验证了该方法。结果表明,该方法有效地降低了向量维数、对未知指纹具有较高的识别率。
To improve the approach of unknown fingerprint-corresponding operating system, this paper presents a classification method based on SVD（singular value decomposition） and DAG（directed acyclic graph）. Firstly, the matrix generated by the operating system fingerprint is converted into the singular value feature. Then Multi-class classification generated by DAG_SVMS is used to judge the unknown fingerprint. Finally the method in the Nmap fingerprint database is verified. The experimental results indicate that this method can effectively reduce the vector dimension and has fairly high recognition rate of the unknown fingerprint.
出处
《信息安全与通信保密》
2013年第9期66-67,72,共3页
Information Security and Communications Privacy
关键词
操作系统识别
NMAP
操作系统指纹
奇异值特征
有向无环图
operating system recognition
Nmap
operating system fingerprint
singular value feature
directed acyclic graph
