特征融合的射频指纹识别方法  

RF fingerprinting identification with feature fusion

作　　者：刘燕平[1] 田金鹏[1] 陈泳[1] 

机构地区：[1]上海大学通信与信息工程学院,上海200444

出　　处：《上海大学学报（自然科学版）》2017年第3期408-413,共6页Journal of Shanghai University:Natural Science Edition

基　　金：国家自然科学基金重点项目面上资助项目(61171086);上海大学创新基金资助项目(sdcx2012041)

摘　　要：在射频指纹(radio frequency fingerprint,RFF)识别系统中,考虑到同一发射机的鲁棒性与不同发射机之间的差异性,提出了将瞬态信号二阶谱中的功率谱密度和互功率谱密度两个特征融合作为指纹的方法,并结合径向基概率神经网络分类器来进行分类.同时,对同一型号两个系列的多种无线网卡进行了分类检测,并与不同的特征提取方法和分类器进行了比较.结果表明,与已有方法相比,此方法的分类精确度有较大的提高.Considering the intra-robustness and inter-difference of transmitters in a radio frequency fingerprinting （RFF） identification system, this paper fuses the second-order spectra, i.e., power spectral density and cross-power spectral density of signals as fingerprints, and uses a radial basis probabilistic neural network as the classifier. The classification performance of the wireless network in two different series has been evaluated in simulation experiments. Compared with other feature extraction methods and classifiers, it is demonstrated that accuracy of the proposed method makes a great improvement.

关 键 词：射频指纹识别 二阶谱 径向基概率神经网络 

分 类 号：TN918[电子电信—通信与信息系统]