基于神经网络学习的网络防火墙拦截效能评估

Network firewall interception effectiveness evaluation based on neural network learning

为了提高对网络防火墙拦截效能智能评估能力,提出一种基于网络攻击特征大数据采样和卷积神经网络学习的网络防火墙拦截效能智能评估方法。采用攻击强度评估方法进行防火墙的拦截能力评估,结合卷积神经网络学习方法进行攻击大数据特征提取,以入侵信息拦截的概率为评价对象,结合定量递归分析方法进行网络防火墙拦截效能自适应动态评估。仿真测试结果表明,该模型能有效评估网络防火墙的拦截效能,提高对入侵信息的拦截检测能力。

In order to improve the capability of intelligent evaluation of network firewall interception effectiveness, an intelligent evaluation method based on big data sampling and convolution neural network learning is proposed. The attack intensity evaluation method is used to evaluate the interception ability of firewall, and the feature extraction of attacking big data is carried out by combining the learning method of convolutional neural network. The probability of intrusion information interception is regarded as the evaluation object. Combined with quantitative recursive analysis method, adaptive dynamic evaluation of network firewall interception effectiveness is carried out. The simulation results show that the model can effectively evaluate the interception efficiency of network firewall and improve the interception detection capability of the intrusion information.