网络病毒感染下振荡攻击数据混淆分离算法

Oscillating Attack Data Confussion Separation Algorithm in Network Virus Infection

通过对网络病毒感染下振荡攻击数据混淆分离,实现对网络病毒攻击数据的特征定位和提取,有效检测网络病毒数据。传统方法中对网络病毒感染下的振荡攻击数据的检测和分离方法使用行为特征分析和卡尔曼滤波方法,算法受到数据振荡和线性特征干扰的影响,检测性能不好。提出一种基于双线性本征波匹配的振荡攻击数据混淆分离算法,构建网络病毒感染振动攻击模型,提取攻击数据的信号模型特征,根据调解病毒样本序列的线性化程度,确定双线性本征陷波器频率参数和带宽参数,实现对信号的滤波,进而实现混淆分离算法的改进。仿真结果表明,该算法设计的滤波器进行振荡攻击数据的混淆分离,具有较好的抗噪能力和干扰抑制能力,对网络病毒攻击数据的混淆分离均方根误差较小,对病毒感染下的振荡攻击信号的检测性能较高,展示了其优越性能。

Through the oscillating attack data obfuscation separation of network virus infection, realize the feature location and extraction of network virus attack data, effective detection of network virus data. Detection and separation method for data of network oscillations attack viral infection in the traditional method using behavior analysis and Calman filtering method, the algorithm is affected by the data oscillation and linear characteristics of interference, the detection performance is not good. To solve this problem, proposed a bilinear intrinsic wave matching oscillation attack data obfuscation separa?tion algorithm based on the vibration model of network virus infection in constructing attack, attack data signal model of fea?ture extraction, and then the bilinear intrinsic wave matching filter design based on the membership function of the fuzzy re?lation and construction of virus infection, achieve the signal filtering, so as to realize the improvement of confusion separa?tion algorithm. Simulation results show that the algorithm is designed to confuse the filter separation of data oscillation at?tack, inhibition ability has better competence of anti noise and interference, to confuse the separation of network virus at?tack data, root mean square error is smaller, higher performance detection on the oscillation signal to attack the virus infec?tion, demonstrating its superior performance.