基于射频指纹的测控地面站身份识别方法  

Radio frequency fingerprint-based TT&C ground station identification method

作　　者：崔天舒 赵文杰 黄永辉 张紫晗 安军社 Cui Tianshu;Zhao Wenjie;Huang Yonghui;Zhang Zihan;An Junshe(Key Laboratory of Electronics and Information Technology for Space Systems,National Space Science Center,Chinese Academy of Sciences,Beijing 100190,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院国家空间科学中心复杂航天系统电子信息技术重点实验室,北京100190 [2]中国科学院大学,北京100049

出　　处：《航天电子对抗》2021年第3期6-9,23,共5页Aerospace Electronic Warfare

摘　　要：卫星测控信道具有开放性特点,第三方容易截获测控信号,并通过特定信号分析手段获得通信编码体制甚至密钥,严重危害测控安全。为增强卫星的测控安全性,提出了一种基于射频指纹的测控地面站身份识别方法,即提取测控信号的射频指纹来验证测控指令是否来自于合法用户。使用卷积神经网络从21台发射机的变功率信号中识别特定发射机,识别精度高达86%,说明可通过射频指纹有效识别发射机身份,提高测控安全性能。Due to the broadcasting characteristics of satellite TT&C channels,TT&C signals are easily intercepted by the adversary,and side-channel attack techniques are used to obtain communication coding systems and even keys,which bring great harm to the security of satellite communication security.In order to enhance satellite TT&C security,a radio frequency fingerprint-based TT&C signal specific emitter identification method to verify whether the command comes from a legitimate user is proposed.A convolutional neural network is used to identify specific emitter from the variable power signals of 21 transmitters,and the recognition accuracy is higher than 86%.It shows that the transmitter’s identity can be effectively identified through radio frequency fingerprints,and TT&C security can be improved.

关 键 词：测控安全 射频指纹 身份识别 卷积神经网络 深度学习 

分 类 号：TN975[电子电信—信号与信息处理]