摘要： 网络安全日志数据库是一种历史数据，对它的分析具有十分重要的实际价值，作为一种时序数据库，针对它的信息挖掘已研究出许多方法。该文提出了一种新的对此类时序数据库的信息挖掘方法，利用小波变换多分辨率分析的方法对信号化后网络安全日志数据库中的数据在不同的时间尺度上进行分析和信息挖掘，从中提取出单位时间内网络受到攻击次数的时间周期规律，并对这种方法的分析特性进行了阐述，而且利用小波阈值重建的方法对原始信号数据进行去噪处理，收到了良好的效果。

关键词: 小波变换, 信号化, 网络安全日志数据库, 数据挖掘

Abstract: Network security log file database is a kind of historical database. It is very important to research it. A lot of data mining methods to research it are found. A new method is provided to analyze and mine this kind of time serial database. Multi-resolution analysis of wavelet transform method is used to analyze and mine the data of network security log file database on different time scale. The period law of the attack number every hour is found by this method. Finally the wavelet threshold method is used to de-noise from the data. It is proved that this method can get good results.

Key words: Wavelet transform, Signalizing, Network security log file database, Data mining