多径传输环境下的射频指纹估计方法

Radio Frequency Fingerprint Estimation in Multi-path Transmission Environment

随着无线通信技术的飞速发展,其安全问题日益严峻.物理层安全技术利用无线信号的内在属性建立安全机制,为增强无线通信安全提供了新思路.无线信道密钥生成技术利用无线信道的互易性、随机性和唯一性生成对称密钥;射频指纹(Radio Frequency Fingerprint, RFF)识别技术借助寄生在无线信号中的射频特征进行设备身份认证.但是,在接收到的信号中,无线信道和RFF混叠在一起,不利于密钥生成和设备识别.无线信道与RFF的分离是物理层安全技术中需要解决的一个重要问题.本文研究了多径传输环境下的RFF估计方法,提出了基于信道互易性(Channel Reciprocity based, CR)的RFF估计方法以及基于主路径分解(Main Path Decomposition based, MPD)的RFF估计方法. CR方法利用无线信道的互易性获取发射机RFF,可广泛适用于单天线或多天线时分双工系统;MPD方法利用接收机的多天线优势,通过重构信号子空间来获取主路径中包含的发射机RFF.仿真结果表明, CR方法可以在理想的高信噪比条件下获得较好的估计性能, MPD方法可以从主路径中分解得到发射机RFF,可有效降低噪声的影响,在低信噪比条件下仍具有良好的估计性能.

With the rapid development of wireless communications technology, the corresponding security problem is becoming more and more serious. Physical layer security technology uses the intrinsic attributes of wireless signals to establish security mechanism, which provides a new idea for enhancing wireless communication security. Wireless channel key generation technology uses reciprocity, randomness and uniqueness of wireless channel to generate symmetric keys, Radio Frequency Fingerprint(RFF) identification technology uses radio frequency features being inherent in the wireless signal to authenticate an equipment. However, the received signal is the combination of wireless channels and RFF, which brings interference to key generation and device identification. Therefore, the separation of wireless channels and RFF plays a key role in physical layer security. This paper studies the RFF estimation method in multi-path transmission environment, and proposes a channel reciprocity based(CR) RFF estimation method and a main path decomposition based(MPD) RFF estimation method.The CR method utilizes the reciprocity of wireless channel to obtain RFF of a transmitter, which can be widely used in single antenna or multi-antenna time division duplex systems. The MPD method utilizes the multi-antenna advantage of a receiver to obtain RFF of a transmitter in the main path by reconstructing the signal subspace. Simulation results show that the CR method can obtain good estimation performances under ideal high signal to noise ratio conditions, the MPD method can decompose the transmitter RFF from the main path, which can effectively reduce the impact of noise,and still has good estimation performance under low signal to noise ratio conditions.