A Novel Optical Transmitter With Chaotic Fingerprint for Identity Authentication in Physical-Layer Security of Optical Networks

作者：Pengjin Zhu, Hongxiang Wang, Yuefeng Ji · 发表于：IEEE Internet of Things Journal · 年份：2024 · DOI：10.1109/jiot.2024.3471853 · 被引用次数：4 · 研究领域：Fractal and DNA sequence analysis、Biometric Identification and Security、Chaos-based Image/Signal Encryption

In this article, we propose and experimentally demonstrate an optical transmitter system with chaotic fingerprint for identity authentication in physical-layer of optical networks. The construction and identification of chaotic fingerprints are introduced in detail. The illegal optical transmitter can be detected through fingerprint identification. We compared the recognition performance of five neural networks (NNs), among which 2D-CNN has the best performance. The simulation results show that the recognition accuracy can reach 100%, and the initial values will not affect the accuracy. Moreover, the experimental results show that the recognition accuracy of legal transmitters is 99.2%, and the detection accuracy of illegal transmitters is 95.7%. Compared to other fingerprint schemes, chaotic fingerprints have higher security, larger fingerprint space, stronger anti-noise ability, and better flexibility. However, the chaotic fingerprint and user signal are separated. To prevent illegal attackers from stealing fingerprints, we use fractional Fourier transform (FRFT) for encryption. The FRFT module can not only achieve the secure connection between the chaotic fingerprint and the user signal but also realize the concealment of time-delay signatures (TDSs) of the fingerprint, which ensures the security of fingerprints. To sum up, the proposed system can resist eavesdropping and injection attacks simultaneously, which provides security for optical networks.