基于粗糙集和SPSO的网络入侵检测方案

A Network Intrusion Detection Scheme Based on Rough Set and Simple Particle Swarm Optimization

针对计算机网络中的安全性问题,提出一种基于粗糙集和简化粒子群优化（SPSO）的网络入侵检测方案。首先,利用粗糙集理论从入侵数据集中提取出分类效果较好的简约特征集。然后,利用训练数据训练BP神经网络分类器,并利用改进后的SPSO优化神经网络的权值和阈值参数。最后,以提取的特征为输入,利用优化后的BP神经网络进行网络入侵分类。在DARPA数据集上进行实验,结果表明该方案能够精确的检测U2R、R2L、DoS和PRB类网络攻击,整体分类准确率达到了87%。

For the security problem of the computer network, a network intrusion detection scheme based on rough set and simplified particle swarm optimization is proposed. Firstly, the rough set theory is used to extract the simple feature set with good classification effect from the intrusion data. Then, the training data is used to train the BP neural network classifier, and the weights and threshold parameters of the neural network are optimized by the improved SPSO algorithm. Finally, the extracted features are regarded as inputs of the optimized BP neural network, so as to realize the classification of network intrusion. Experiments on DARPA data sets show that the proposed scheme can accurately detect U2 R, R2 L, DoS and PRB attacks, and the overall classification accuracy rate reaches 87 %.