引用本文
  • 徐锦龙,魏冬,黄伟庆.基于极化指纹的辐射源个体识别方案[J].信息安全学报,已采用    [点击复制]
  • Xu Jinlong,Wei Dong,Huang Weiqing.Individual Emitter Identification Scheme Based on Polarization Fingerprint[J].Journal of Cyber Security,Accept   [点击复制]
【打印本页】 【下载PDF全文】 查看/发表评论下载PDF阅读器关闭

过刊浏览    高级检索

本文已被:浏览 27043次   下载 26768  
基于极化指纹的辐射源个体识别方案
徐锦龙1,2, 魏冬1,2, 黄伟庆1,2
0
(1.中国科学院信息工程研究所第四研究室 北京 中国;2.中国科学院大学网络空间安全学院 北京 中国)
摘要:
对于资源有限的个体辐射源,射频指纹(Radio Frequency Fingerprint, RFF)是一种低成本、高效率以及高安全性的识别技术。然而目前RFF面临着指纹稳定性低、应用难度大的问题。为了解决这些问题,本文首次提出极化指纹的概念以及对应的辐射源个体识别方案。首先通过对经典圆极化贴片天线形成圆极化的过程进行分析得到极化指纹的数学模型。极化指纹来源于天线的结构和硬件缺陷在信号极化中留下的包含身份信息的特征,这种特征表现为信号极化状态的频率相关性。同时本文对极化指纹的群体特征和个体特征进行分析。群体特征表征天线的结构信息,其尺度更大,因此抵抗噪声影响的能力更强。个体特征表征辐射源身份信息,其尺度更小,因此抵抗噪声的能力更弱。基于这两个特征设计的两级特征模板极化指纹库和两步识别算法使得该方案在面对大量设备时能够保持较高的识别效率。由于目前的通信系统不会对极化进行调制,相比于RFF,极化指纹可以稳定持续存在,不仅使得该方案更容易实现,还可以获得更多的采样量。更重要的是,通过推导该方案的误警率和准确率证明了提高采样量可以降低噪声对极化指纹造成的失真。最后基于无线物联网(IoT)设备进行实验,实验结果表明在相同条件下,基于极化指纹的辐射源个体识别方案比基于RFF的方案有更佳的性能。
关键词:  物理层安全  电磁指纹  极化指纹  辐射源个体识别
DOI:10.19363/J.cnki.cn10-1380/tn.2023.09.19
投稿时间:2021-11-23修订日期:2021-12-20
基金项目:
Individual Emitter Identification Scheme Based on Polarization Fingerprint
Xu Jinlong1, Wei Dong1, Huang Weiqing2
(1.The 4th Laboratory,Institute of Information Engineering,Chinese Academy of Sciences;2.The th Laboratory,Institute of Information Engineering,Chinese Academy of Sciences)
Abstract:
For individual emitter with limited resources, Radio Frequency Fingerprint (RFF) is a low cost, high efficiency and high security identification technology. However, currently RFF faces the problems of low fingerprint stability and high application difficulty. In order to solve these problems, we propose an individual emitter identification scheme based on polarization fingerprint (PF). We analyzed the formation of the circular polarization via classic circularly polarized patch antennas and constructed a mathematical model of PF. PF originates from the feature that contains identity information left by the structure and the hardware imperfections of the antenna in the signal polarization, and this feature is reflected in the frequency dependence of the signal polarization state. At the same time, we analyzed the group feature and individual feature of PF. The group feature represents the structural information of the antenna, and its scale is larger, so the ability to resist the influence of noise is stronger. Individual feature represents the identity information of the emitter, and its scale is smaller, so the ability to resist noise is weaker. The two-level template PF database and two-step identification algorithm designed based on these two features enable the scheme to maintain high identification efficiency when facing a large number of devices. Because current communication systems does not modulate the polarization, compared to RFF, PF can exist stably and continuously, which not only makes the scheme easier to implement, but also helps the scheme to obtain more samples. More importantly, by de-riving the false alarm rate and accuracy rate of the scheme, it is proved that increasing the sampling amount can reduce the distortion of PF caused by noise. Finally, we conducted experiments based on wireless Internet of Things (IoT) devices. Results show that under the same condition, the identification scheme based on PF performs better than that based on RFF.
Key words:  physical layer security  electromagnetic fingerprint  polarization fingerprint  individual emitter identification