基于主成分分析和概率神经网络的入侵检测方法

Intrusion Detection Method Based on Principal Component Analysis and Probabilistic Neural Network

针对神经网络在入侵检测的应用中存在入侵数据冗余信息多,数据量大,训练时间长,易陷入局部最优等问题,提出了一种基于主成分分析(PCA)和概率神经网络(PNN)的入侵检测方法。首先使用PCA对数据进行特征降维,解决了入侵数据冗余信息多的问题;然后使用PNN建立入侵检测模型;其次,使用粒子群算法(PSO)解决概率神经网络参数的优化问题;最后使用KDD99数据集对该模型进行测试。实验结果表明:该方法能够有效提高检测的效果,而且检测速度明显提高。

In view of the redundant information,big volume of data and long training time of neural network application in intrusion detection,and easy to get trapped in local optimum,an intrusion detection method based on principal component analysis and probabilistic neural network is proposed in this paper.Firstly,the dimension of the original intrusion data is reduced by principal component analysis,which solves the problem of redundant information of intrusion data.Secondly,the probabilistic neural network is used to establish the intrusion detection model.Then,the problem that it is hard to optimize the parameters of the probabilistic neural network is solved by using particle swarm optimization algorithm.Finally,the KDD99 data set is used to test the performance of the model.The experimental result shows that both the precision rate and detection speed are improved effectively by using the proposed method.