Journal of Lightwave Technology

End-to-End Learning-Driven Dual Independent Key Generation and Distribution for Optical Network Physical Layer Security

Yanwen Zhu, Yixin Wang, Yuang Li, Zirui Ding, Yi Huang, Huibin Zhang, Yajie Li, Yongli Zhao, Jiarui Zhang

Published Jan 13, 2026• 0 citations

Abstract

To address the urgent demand for secure optical fiber communication, a novel dual independent key generation and distribution scheme for Optical Network Physical Layer Security is proposed for the first time. End-to-end (E2E) learning with a Transformer algorithm featuring adaptive positional encoding is introduced to accurately model coherent fiber transmission systems, achieving an R-squared value exceeding 0.95 for waveform fitting and data distribution. The error vector phase (EVP) and amplitude noise (AN) of the E2E system are utilized to comprehensively characterize channel uniqueness, reducing system reconstruction risks while serving as sources for highly random key sequences for legitimate parties. Key consistency stems from channel reciprocity and precise E2E modeling, with feasibility analyzed across diverse link conditions. Leveraging the fact that eavesdroppers cannot access the system's full lifecycle data, keys are extracted from outputs of transfer learning (TL)-updated E2E models to enhance security. Security analysis under various eavesdropping scenarios demonstrates resistance to attacks. Experimental results show that, with FPGA-implemented post-processing, error-free key distribution is achieved over 120 km SSMF at EVP and AN key rates of 306.74 Mbps and 521.82 Mbps, respectively.

DOI: 10.1109/jlt.2026.3653785

Related papers

Fiber-optic physical layer key generation and distribution based on the phase of the error vector in end-to-end systems

Yanwen Zhu, Yixin Wang, Y. Li +6 more

This paper proposes an end-to-end physical layer secure key generation and distribution (E2E-PLSKGD) scheme based on error vector phase (EVP) for fiber transmission systems. It addresses compatibility issues in existing PLSKGD caused by additional random sources or devices, as well as limitations in application scenarios and insufficient key generation rate (KGR) from amplitude or phase modulation of intensity-modulated signals. We introduce a position-enhanced transformer (PE-transformer) architecture with adaptive position coding (APC) for accurate E2E modeling in coherent optical systems. Leveraging the inherent high randomness of signal phase in fiber systems, legitimate parties use the PE-transformer to build E2E models via bidirectional transmission, compute error vectors, and extract EVP noise as key quantization sequences. Using data from new negotiation cycles and transfer learning (TL), local E2E source models are rapidly updated to target models, enhancing key security. Simulations show superior E2E modeling, with R-Square values >0.95, excellent waveform fitting, and consistent data distributions across distances. Pearson correlation coefficients (CC) of EVP sequences between legitimate parties exceed 0.8 across 50-80 km distances and -2 dBm to 4 dBm launch powers, indicating strong correlation. Experimental validation on a 120 km fiber system shows an EVP sequence correlation of 0.72, with real-time post-processing achieved via FPGA. At a 5G baud rate, error-free KGR reaches 0.8 Gbps. The proposed architecture enables precise E2E modeling, offering a new approach for dynamic, high-quality key generation and a potential method for AI-empowered fiber transmission security.

Optics ExpressJul 24, 20252 citations

Key Distribution Scheme for Optical Fiber Channel Based on SNR Feature Measurement

Xiangqing Wang, Jie Zhang, Bo Wang +4 more

With the increase in the popularity of cloud computing and big data applications, the amount of sensitive data transmitted through optical networks has increased dramatically. Furthermore, optical transmission systems face various security risks at the physical level. We propose a novel key distribution scheme based on signal-to-noise ratio (SNR) measurements to extract the fingerprint of the fiber channel and improve the physical level of security. The SNR varies with time because the fiber channel is affected by many physical characteristics, such as dispersion, polarization, scattering, and amplifier noise. The extracted SNR of the optical fiber channel can be used as the basis of key generation. Alice and Bob can obtain channel characteristics by measuring the SNR of the optical fiber channel and generate the consistent key by quantization coding. The security and consistency of the key are guaranteed by the randomness and reciprocity of the channel. The simulation results show that the key generation rate (KGR) can reach 25 kbps, the key consistency rate (KCR) can reach 98% after key post-processing, and the error probability of Eve’s key is ~50%. In the proposed scheme, the equipment used is simple and compatible with existing optic fiber links.

PhotonicsJun 9, 20214 citations

Secure Key Distribution System Based on Optical Channel Physical Features

Xiangqing Wang, Jie Zhang, Yajie Li +2 more

This paper proposes a novel key distribution system based on the measurement of the bit error rate (BER) in a fiber channel. By carrying out the loopback BER measurement at both the transmitting and the receiving end, the BER is quantified and codified to generate consistency keys. Alice and Bob respectively encrypt the signal with the help of the public key base at the transmitting end. At the receiving end the data are decrypted and then the BER is measured. The security of generated keys in the system is enhanced by the use of the encryption base and the randomness of the channel. Additionally, noise sources are added to the channel so that the random noise conceals signals and Eve cannot eavesdrop the useful information in it. Due to the randomness of the channel noise, the generated keys have good randomness. This system at a high key generation rate is compatible with the existing communication device and its measurement is simple. The 10G bit/s-200 km coherent optical communication system is used to measure the BER of the security feature information of the extracted channel. The experimental results show that the key distribution system obtains a consistency rate of 98% with the key generation rate 2 Mbps, both of which have been significantly improved.

IEEE photonics journalNov 15, 201933 citations

Unidirectional physical layer secure key distribution in a fiber channel assisted by neural networks

Liuming Zhang, Xinran Huang, Zhi Chai +3 more

Physical layer secure key distribution (PLSKD) in fiber networks provides high-level security originating from the channel uniqueness, while the channel reciprocity is shared between the legal users via bidirectional transmission. To ensure the key consistency, perfect wideband bidirectional optical devices are required for high-speed PLSKD. Here, we propose and demonstrate a unidirectional PLSKD scheme, where the channel reciprocity is reproduced by neural networks (NNs) so that highly correlated key waveforms are generated and shared between the legal users. Experimental results show that an error-free key generation rate (KGR) of 3.8 Gb/s with excellent randomness is achieved for an 11-km standard single-mode fiber (SSMF) channel. The proposed scheme alleviates the stringent requirement of bidirectional channel transmission in high-speed PLSKD and enables the integration of PLSKD and data transmission in the same fiber channel.

Optics LettersJul 20, 202217 citations

Physical-layer encryption and authentication scheme based on SKGD and 4D hyper-chaos

Danyang Wang, Hongxiang Wang, Xu He +1 more

In this paper, a scheme to realize encryption and digital identity authentication at the same time is proposed for enhancing the physical-layer security of point-to-point optical links (PPOL). Exploiting identity code encrypted by the key as authentication information effectively resists passive eavesdropping attacks in fingerprint authentication. The proposed scheme theoretically realizes secure key generation and distribution (SKGD) by phase noise estimation of the optical channel and the generation of identity codes with good randomness and unpredictability by the four-dimensional (4D) hyper-chaotic system. The local laser, erbium doped fiber amplifier (EDFA), and public channel provide the entropy source of uniqueness and randomness to extract symmetric key sequences for legitimate partners. The simulation conducted in a quadrature phase shift keying (QPSK) PPOL system over 100km standard single mode fiber verify successfully that 0.95Gbit/s error-free SKGD. The unpredictability and high sensitivity to the initial value and control parameters of the 4D hyper-chaotic system provide a huge space of ~10 125 for identity codes, which is sufficient to resist exhaustive attack. With the proposed scheme, the security level of key and identity can be increased markedly.

Optics ExpressMar 9, 202313 citations

Secure key generation and distribution scheme based on historical fiber channel state information with LSTM

Danyang Wang, Hongxiang Wang, Yuefeng Ji

In this paper, a scheme to realize unclonable physical-layer security key generation and distribution (PL-SKGD) based on historical fiber channel state information (HFCSI) is proposed. PL-SKGD schemes based on channel characteristics for enhancing the physical-layer security of optical networks have been proposed in recent years. However, there are potential disadvantages in these schemes, such as 1) low key generation rate (KGR): the slow frequency of the analog waveform change of the channel characteristic leading to low KGR; 2) incompatibility with existing infrastructure: active scrambling to increase the frequency of channel characteristic changes, or tracking changes of channel characteristics requires additional devices; 3) easy to be cloned: all of the optical channel state information is reflected in the signal transmitted inside the fiber, which makes it easy to reproduce by illegal eavesdropper through features analysis and other methods. In order to solve the above problems, a PL-SKGD scheme is designed which uses the chain structure composed of long short-term memory neural network (LSTM-NN) units to learn and store the unique mapping relationship between historical channel time series and provides unclonability based on the fundamental fact that the eavesdropper Eve can never obtain the full HFCSI. The simulation conducted in a quadrature phase shift keying point-to-point optical link system verified successfully that KGR = 0.82 Gbit/s error-free SKGD. The loss function of LSTM-NN drops sharply in the early stages of training and remains a small value. The security of the SKGD system is analyzed, which effectively improves the unclonability of the system. Finally, it is verified that the optimal fiber channel length for error-free SKGD of the proposed scheme is 150 km considering the error correction capability of information reconciliation and weighing key sequence error rate and valid bit generation rate.

Optics ExpressDec 19, 20237 citations