基于频谱包络特征提取的PUE攻击检测研究  

RESEARCH ON PUE ATTACK DETECTION BASED ON SPECTRAL ENVELOPE FEATURE EXTRACTION

作　　者：刘云飞[1] 郑文秀[1] 刘悦凯 

机构地区：[1]西安邮电大学通信与信息工程学院,陕西西安710061

出　　处：《计算机应用与软件》2016年第11期277-281,共5页Computer Applications and Software

基　　金：国家自然科学基金项目(61301091)

摘　　要：提出一种基于频谱包络特征提取的PUE(Primary User Emulation)攻击检测方法。在论证频谱包络起伏特征可以作为指纹特征提取的基础上,结合曲线拟合,选取特征参数,构建能够明显反映频谱包络起伏特征的向量,通过FCM聚类区分主用户和PUE攻击用户。仿真实验表明,该方法能够有效区分主用户和PUE攻击用户,具有较好的可行性和可靠性。This paper presents a PUE( Primary User Emulation) attack detection method,which is based on spectral envelope feature extraction. On the basis of having demonstrated that the fluctuant features of spectral envelope can be extracted as the fingerprint features,we combine the curve fitting and select the characteristic parameters,and construct the vector which can obviously reflect the fluctuation characteristic of spectral envelope,then we differentiate the primary user from PUE attack user by fuzzy c-means clustering. Simulation experimental results show that this method can effectively distinguish primary user and PUE attack user,and has good feasibility and reliability.

关 键 词：PUE攻击 辐射源 频谱包络 特征提取 曲线拟合 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]