基于人工蜂群算法的网络攻击流量自动辨识研究

Research on Automatic Identification of Network Attack Traffic Based on Artificial Bee Colony Algorithm

传统的网络攻击流量自动辨识方法在网络流量较大情况下的寻优效果差,导致攻击流量识别性能降低,笔者提出了基于人工蜂群算法的网络攻击流量自动辨识方法。该方法在网络正常运行过程中进行流量采样,重新设计网络协议模型结构,获取攻击流量多个特征集,利用方差矩阵筛选权重较大的特征集作为判断依据,利用人工蜂群算法对模糊神经网络的隶属度函数中心及宽度,提升寻优效果,设计出整体的网络攻击流量自动辨识流程。为验证方法有效性,设计模拟实验与2种传统方法进行对比,实验结果表明,在相同识别率条件下,设计方法的误识别率更低,在不同类型的攻击流量测试下,设计方法的识别效果明显优于2种传统方法。

The traditional automatic identification method of network attack traffic has poor optimization effect under the condition of large network traffic,which leads to the degradation of the identification performance of attack traffic.An automatic identification method of network attack traffic based on artificial bee colony algorithm is proposed.During the normal operation of network,the traffic is sampled and the network protocol model structures are redesigned and the attack traffic multiple characteristics are obtained based on the variance matrix filter weight larger feature set as the basis for judgment.Artificial bee colony algorithm is used to improve the optimization effect of fuzzy neural network membership function center and width,and the whole automatic identification process of network attack traffic is designed.In order to verify the effectiveness of the method,a simulation experiment was designed to compare with the two traditional methods.The experimental results show that the false recognition rate of the design method is lower under the condition of the same recognition rate,and the recognition effect of the design method is obviously better than that of the two traditional methods under different types of attack traffic test.