一种基于禁忌神经网络的网络入侵检测模型

随着互联网的发展和普及,传统网络入侵防范方法如防火墙、数据加密等已经很难保证系统和网络资源的安全。为此,本文设计了基于改进禁忌算法和神经网络的网络入侵检测方法。首先建立三层的BP神经网络模型用于实现入侵检测。然后通过BP反向传播算法获取网络的权值和阀值等参数,并设计了一种基于双禁忌表的改进禁忌优化算法,采用此改进的禁忌优化算法对BP算法优化得到的权值和阀值进行进一步寻优。最后,将禁忌算法优化后的神经网络用于网络入侵检测。仿真实验表明,此方法能够有效地实现网络入侵检测,具有较快的收敛速度和较高的检测率,是一种适合网络入侵检测的可行方法。

A Network Intrusion Detection Model Based on Tabu Neural Network

With the further development of the Internet and the popularization of network,the traditional network detection methods such as firewall and data encryption cannot guarantee the security of system and network resource.Therefore,the network intrusion method based on the improved tabu algorithm and neural network is proposed in this paper.Firstly,the three-layer BP model for network intrusion detection is set up,then the BP back propagation algorithm is used to obtain the parameters such as weight and threshold,and an improved tabu algorithm based on double tabu table is designed and used to optimize the parameters such as weight and threshold.Finally,the tabu algorism optimizing neural network is put forward to detect network intrusion.The simulation experiments show that this method can effectively realize the network intrusion detection.The method has faster convergence speed and higher detection rate,so it is suitable for a feasible way to detect network intrusion.