摘要
目前,网络入侵技术越来越先进,许多黑客都具备反检测的能力,他们会有针对性地模仿被入侵系统的正常用户行为;或将自己的入侵时间拉长,使敏感操作分布于很长的时间周期中;还可能通过多台主机联手攻破被入侵系统.对于伪装性入侵行为与正常用户行为来说,仅靠一个传感器的报告提供的信息来识别已经相当困难,必须通过多传感器信息融合的方法来提高对入侵的识别率,降低误警率.应用基于神经网络的主观Bayes方法,经实验,效果良好.
Nowadays, more and more hackers intrude the network successfully by imitating legal users＇ behavior; lengthening the intrusion time or making the intrusion in a working group. It＇s difficult to identify the intrusion by the information gathered through one sensor, so we should use the information fusion method to improve the detection rate and lower the false-alarm rate. The experimental data proves that subjective-Bayes method based on the neural network is efficient in the network intrusion detection.
出处
《海军工程大学学报》
CAS
北大核心
2006年第3期103-107,共5页
Journal of Naval University of Engineering
关键词
信息融合
入侵检测
主观BAYES方法
神经网络
information fusion
intrusion detection
subjective-Bayes method
neural network
