基于门限联合判别的同质网络入侵检测仿真

动态同质网络在路由转发控制中节点分布具有动态性,容易受到病毒入侵,提出基于上下门限联合判别的动态同质网络入侵检测算法。以时间均值和谱密度为网络入侵检测的统计特征量,采用宽平稳随机序列分析方法构建网络入侵的统计信号分析模型,对入侵信号采用相关性检测和同态匹配滤波方法进行降噪和盲源分离处理,结合极速学习方法进行动态同质网络的入侵特征量提取,采用上下门限联合判别方法实现动态同质网络的入侵检测。仿真结果表明,采用该方法进行动态同质网络的入侵检测的准确性较高,抗干扰能力较强,对入侵信息的准确检测概率提升效果显著。

Simulation of Homogeneous Network Intrusion Detection Based on Threshold Joint Discrimination

Dynamic homogeneous network has dynamic distribution of nodes in routing and forwarding control,which is vulnerable to virus intrusion.An intrusion detection algorithm for dynamic homogeneous network based on upper and lower threshold joint discrimination is proposed.The statistical signal analysis model of network intrusion is constructed by using the time mean and spectral density as the statistical characteristic of network intrusion detection and the method of wide-stationary random sequence analysis.Correlation detection and homomorphism matched filter are used for noise reduction and blind source separation of intrusion signals,and dynamic homogeneous network intrusion feature extraction is combined with extreme learning method.The upper and lower threshold discriminant method is used to realize the intrusion detection of dynamic homogeneous network.The simulation results show that the proposed method has high accuracy and strong anti-interference ability for dynamic homogeneous network intrusion detection,and it can improve the probability of intrusion detection significantly.