Due to device-specific defects introduced during the hardware manufacturing process, the radio frequency fingerprint (RFF) can be extracted to identify wireless devices and further avoid spoofing attacks. Many effective RFF identification methods have been proposed based on either machine learning or deep learning. However, from the perspective of communication security, if the RFF of the transmitter can be easily extracted and identified, attackers can disguise themselves as legitimate transmitters by impersonating RFF and other means, thereby undermining the security of wireless communications. Therefore, concealing the RFF of legitimate transmitters from detection and camouflage attacks has become a highly challenging issue in the field of wireless communications. This article presents an active RFF concealment (RFFC) method, which removes the nonlinear features of the transmitter system, thereby preventing attackers from obtaining the transmitter's RFF and ensuring the identity security of the transmitter. To evaluate the performance of RFF concealing technology, we simulate seven types of RFFC systems, and collect datasets without and with RFFC technology. The simulation results show that the performance of traditional transmitter identification methods decreases sharply after RFFC. Especially in low signal-to-noise ratio environments and complex multipath channel conditions, the proposed RFFC technology makes the RFF features chaotic and difficult to detect, leading to dramatically reduced effectiveness of existing transmitter identification methods.

中文翻译：

---

射频指纹隐藏通信安全新前沿


由于硬件制造过程中引入的设备特定缺陷，可以提取射频指纹（RFF）来识别无线设备并进一步避免欺骗攻击。基于机器学习或深度学习，人们提出了许多有效的 RFF 识别方法。但从通信安全的角度来看，如果发射机的RFF能够被轻易提取和识别，攻击者就可以通过冒充RFF等手段，将自己伪装成合法的发射机，从而破坏无线通信的安全。因此，隐藏合法发射机的RFF免受检测和伪装攻击已成为无线通信领域极具挑战性的问题。本文提出一种主动RFF隐藏(RFFC)方法，消除发射机系统的非线性特征，从而防止攻击者获取发射机的RFF，保证发射机的身份安全。为了评估 RFF 隐藏技术的性能，我们模拟了七种类型的 RFFC 系统，并收集不使用和使用 RFFC 技术的数据集。仿真结果表明，传统发射机识别方法在RFFC之后性能急剧下降。特别是在低信噪比环境和复杂的多径信道条件下，所提出的RFFC技术使得RFF特征变得混乱且难以检测，导致现有发射机识别方法的有效性大幅降低。