摘要
由于现有的检测方法准确程度水平低,召回程度在85%以下,为此研究基于小波包的即时通信网络安全漏洞检测方法。该方法通过对通信网络IP特征进行分析并计算聚类中心,实现漏洞数据的聚类。同时,利用小波包滤波进行特征提取和最优分解层数确定,通过相空间重构得到最优去噪结果。针对即时通信网络节点,计算安全漏洞弧长,并应用蚁群搜索算法精确定位漏洞位置。实验结果表明,该方法在样本数为6000个时,准确度达到90%以上,同时召回率超过85%,具备较好的漏洞检测能力。
Due to the low accuracy level of the existing detection methods,the recall degree is be-low 85%,so the detection mcthod based on wavelet packet is studied.This method realizes the clustering of vulnerability data by analyzing the IP features of the communication network and calculating the clustering center.Meanwhile,the wavelet packet filtering is used for feature ex-traction and the optimal decomposition layer number determination,and the optimal denoising results are obtained through the phase space reconstruction.For the instant messaging network nodes,the arc length of the security vulnerability is calculated,and the ant colony search algo-rithm is applied to accurately locate the vulnerability location.The experimental results show that the accuracy of this method is more than 90%when the number of samples is 6000,and the recall rate is more than 85%,which has good vulnerability detection ability.
作者
刘艳
YAN Liu(Hainan Vocational University,Modem Supply Chain College,Haikou,Hainan,570100)
出处
《长江信息通信》
2024年第7期175-177,共3页
Changjiang Information & Communications
关键词
小波包
即时通信
网络
漏洞
检测方法
wavelet packet
instant communication
network
vulnerability
detcction method
