摘要
研究网络安全问题,提高入侵检测效率,针对网络入侵检测传统采用RBF神经网络方法在网络入侵中由于初始权值设定不当导致检测入侵耗时长、正确检测率低,误报和漏报率记的难题,为了解决上述问题,提出了一种GARBF神经网络入侵检测模型。GARBF神经网络模型在网络入侵检测过程中,采用遗传算法对RBF神经网络初始权值进行优化,然后将网络入侵数据输入优化的RBF神经网络中进行学习和检测。结果表明,相比较传统网络入侵检测模型,网络入侵检测误报率、耗时都较低,证明提高网络入侵检测的正确性和效率。
Researching network intrusion detection problem.The initial value set of traditional neural network is random,which may produce incorrect initial weights set and lead the training of traditional neural network algorithm time-consuming and the detection rate is not high.This paper puts forward a hybrid model of neural network optimized by genetic algorithm,using genetic algorithm to optimize the parameters of RBF neural network,putting the optimized results directly into the RBF neural network training,and at last,using the optimized parameters to test the data of RBF neural network.Compared with traditional BP neural network and RBF neural network,the experimental results showed that the GARBF neural network detection rate is higher,and time-consuming is lower,which explained that the RBF neural network optimized by genetic algorithm is effective and feasible.
出处
《计算机仿真》
CSCD
北大核心
2011年第6期165-168,共4页
Computer Simulation
关键词
遗传算法
神经网络
优化
入侵检测
Genetic algorithm
Neural network
Optimization
Intrusion detection