论文部分内容阅读
为了解决当前无线传感器网络安全问题,针对入侵类型多样性,首先将粗糙集引入到无线传感器网络特征约简中,发现特征数据之前的关系,消除特征集中无相关或者影响较小特征,减少分类器输入向量数,降低计算复杂度,然后采用支持向量机对无线传感器入侵检测进行非线性建模与分类,并采用狼群算法对分类器的参数进行优化和选择,最后采用具体数据对算法的性能进行检验。实验结果表明,本文算法提高了无线传感器网络安全性,获得了较高的无线传感器网络入侵检测率,降低了误警率,增强了无线传感器网络防御各种攻击的能力。
In order to solve the current wireless sensor network security problem, aiming at the diversity of intrusion types, the rough set is first introduced into the WSN feature reduction, the relationship between the feature data is found out, the correlation feature of the feature set is eliminated or the feature is not affected, The number of input vectors is reduced, the computational complexity is reduced, and then SVM is used to model and classify wireless sensor intrusion detection. Wollastonian algorithm is used to optimize and select the parameters of the classifier. Finally, the performance of the algorithm To test. The experimental results show that the proposed algorithm improves the security of wireless sensor networks and achieves a higher rate of intrusion detection of wireless sensor networks, reduces the false alarm rate and enhances the capabilities of wireless sensor networks against various attacks.