基于双谱LLE降维的PUE攻击检测

PUE Attack Identification Based on Improved Bi-spectrum Dimension Reduction Using LLE

针对认知无线电中的主用户仿冒攻击检测问题,给出了基于矩形积分双谱的局部线性嵌入降维算法,用于识别主用户(PU)和仿冒用户(SU)。选择矩形积分双谱作为识别特征参数,利用局部线性嵌入算法(LLE)进行特征数据约简,通过基于核函数的支持矢量机(SVM)进行个体识别。实验结果表明,该方法具有较高的识别率,并能够较好解决PUE攻击检测问题。

To address the issue of primary user emulation attack identification in cognitive radio, an improved square integral bi-speetrum dimension reduction method using locally linear embedding algorithm is proposed for the identification of primary user（PU） and secondary user（ SU ）. Square integral hi-spectrum （ SIB ） is used as the feature parameter, and local linear embedding （LLE） is utilized to reduce the feature dimension,then a support vector machine（SVM） based on kernel-function is used for individual identifica- tion.The experimental results demonstrate that the suggested technique has a high recognition rate,and it can solve the problem of PUE attack identification.