Intelligent transportation and autonomous driving require accurate and secure ranging. Quantum entanglement-based ranging provides physical-layer security and picosecond-level synchronization but remains vulnerable to signal sparsity and noise in dynamic environments. We propose a predictive reconstruction-based entangled photon ranging system that improves robustness and accuracy under low signal-to-noise ratio conditions. The system combines adaptive signal preparation, feature-aware processing, and high-fidelity sequence reconstruction through a triple-stream dual-channel differential-sequence network (TDDN) that extracts distribution features from photon arrival sequences. A coincidence-counting model incorporating atmospheric effects is also developed. Across 100 test subsets, the TDDN achieves a mean absolute error of 2.311 cm, a mean absolute percentage error of 2.976%, and a Pearson correlation coefficient of 0.99, reducing the mean absolute error by 18.0%–56.1% relative to multilayer perceptron, gated recurrent unit, long short-term memory, and recurrent neural network baselines under low signal-to-noise ratio conditions. These results demonstrate the potential of the proposed TDDN for robust and scalable ranging in safety-critical applications.
Intelligent transportation and autonomous driving require accurate and secure ranging. Quantum entanglement-based ranging provides physical-layer security and picosecond-level synchronization but remains vulnerable to signal sparsity and noise in dynamic environments. We propose a predictive reconstruction-based entangled photon ranging system that improves robustness and accuracy under low signal-to-noise ratio conditions. The system combines adaptive signal preparation, feature-aware processing, and high-fidelity sequence reconstruction through a triple-stream dual-channel differential-sequence network (TDDN) that extracts distribution features from photon arrival sequences. A coincidence-counting model incorporating atmospheric effects is also developed. Across 100 test subsets, the TDDN achieves a mean absolute error of 2.311 cm, a mean absolute percentage error of 2.976%, and a Pearson correlation coefficient of 0.99, reducing the mean absolute error by 18.0%–56.1% relative to multilayer perceptron, gated recurrent unit, long short-term memory, and recurrent neural network baselines under low signal-to-noise ratio conditions. These results demonstrate the potential of the proposed TDDN for robust and scalable ranging in safety-critical applications.
Recently, ranging has played an increasingly important role in many scenarios, but the problems of limited security and accuracy have always been troubling. The entanglement ranging method provides picosecond-level clock synchronization and unconditional security and is deemed a good solution to meet the growing demand for accuracy and security, promising to play a significant role in the next generation of navigation. This paper focuses on the ranging system based on entangled photons, and proposes solutions to the challenges of accuracy and robustness. Specifically, a coincidence counting calibration algorithm is proposed to suppress the background noise, and a novel data selection algorithm based on relative quantum efficiency is proposed to suppress abnormal coincidence counting. Then, to improve the ranging accuracy even in challenging scenarios, the paper proposed a novel density-based hierarchical filtering algorithm to detect and filter outliers, resulting in more precise and reliable ranging. Finally, extensive experiments are carried out on the built test platform to verify the safety and effectiveness of the proposed system. The results show that the ranging error is 0.0225 m in the test time of 1 ns. The proposed algorithm reduced the ranging error of abnormal data from 1.5205 m to 0.0100 m, and achieved the probability (90%) of ranging error within 0.0150 m.
Quantum technologies, leveraging the principles of quantum mechanics, are rapidly advancing and offer transformative benefits in secure communication, computation, and sensing. While quantum communication systems provide unprecedented data security, the implementation security of quantum communication protocols still needs investigation. In this regard, the security of Quantum Key Distribution (QKD) is well studied, while there is limited research on quantum communication protocols beyond QKD. In this work, we study time-delay attacks on generic entanglement distribution quantum communication networks. These networks rely heavily on precise time synchronization between detection stations and a preset coincidence window for accurate identification of distributed entangled pairs. An eavesdropper can exploit this reliance on correct photon arrival timing by introducing a time delay, causing genuine entangled pairs to fall outside the coincidence window. This leads to missed detections, potential false pair identifications, or denial-of-service (DoS) conditions. To mitigate this vulnerability, we propose photon arrival-time monitoring and threshold-based time-delay detection, enabling adaptive coincidence window adjustment. By dynamically centering the coincidence window on the mean photon arrival time, valid entangled pairs can still be accurately identified, even in the presence of a time-delay attack. We simulate the quantum measurement process under different time-delay insertion values in the quantum channel, assuming perfectly synchronized remote detection stations. Our results show the impact of time-delay attacks on entangled pair detection rates, detection accuracy, and fidelity of the reconstructed Bell state. The effectiveness of our proposed countermeasure is demonstrated in simulation, showing that it maintains both the photon count rate and the fidelity at high levels, even when an eavesdropper introduces time delays in the quantum channel.
Quantum technologies, leveraging the principles of quantum mechanics, are rapidly advancing and offer transformative benefits in secure communication, computation, and sensing. While quantum communication systems provide unprecedented data security, the implementation security of quantum communication protocols still needs investigation. In this regard, the security of Quantum Key Distribution (QKD) is well studied, while there is limited research on quantum communication protocols beyond QKD. In this work, we study time-delay attacks on generic entanglement distribution quantum communication networks. These networks rely heavily on precise time synchronization between detection stations and a preset coincidence window for accurate identification of distributed entangled pairs. An eavesdropper can exploit this reliance on correct photon arrival timing by introducing a time delay, causing genuine entangled pairs to fall outside the coincidence window. This leads to missed detections, potential false pair identifications, or denial-of-service (DoS) conditions. To mitigate this vulnerability, we propose photon arrival-time monitoring and threshold-based time-delay detection, enabling adaptive coincidence window adjustment. By dynamically centering the coincidence window on the mean photon arrival time, valid entangled pairs can still be accurately identified, even in the presence of a time-delay attack. We simulate the quantum measurement process under different time-delay insertion values in the quantum channel, assuming perfectly synchronized remote detection stations. Our results show the impact of time-delay attacks on entangled pair detection rates, detection accuracy, and fidelity of the reconstructed Bell state. The effectiveness of our proposed countermeasure is demonstrated in simulation, showing that it maintains both the photon count rate and the fidelity at high levels, even when an eavesdropper introduces time delays in the quantum channel.
Accurate state-of-health (SOH) estimation for lithium-ion batteries remains a challenging problem due to complex electrochemical degradation mechanisms and long-range temporal dependencies. In this work, we propose a quantum-enhanced recurrent framework, termed QLSTM, in which variational quantum circuits are directly embedded into the gating mechanisms of long short-term memory networks. By replacing classical affine transformations with parameterized unitary operations, the proposed model introduces structured nonlinear transformations into the recurrent state-transition process. Extensive experiments on multiple benchmark battery datasets demonstrate that QLSTM consistently outperforms classical sequence models in both predictive accuracy and robustness, achieving significant reductions in mean absolute error (MAE), with improvements on the order of 20% compared with classical LSTM baselines. Ablation studies further confirm that these improvements arise primarily from quantum-enhanced gating rather than input-level transformations. Additional analyses on qubit scaling and noise robustness reveal that model performance is governed by a balance between expressive capacity and trainability. These results provide empirical evidence that embedding quantum computational primitives within recurrent architectures offers a structurally grounded approach to improving sequence modeling capability. The proposed framework establishes a new design paradigm for integrating quantum operators into temporal learning models, with potential applications in complex dynamical system prediction tasks.
Synchronization is a fundamental requirement for the reliability and security of entanglement-assisted quantum key distribution (QKD), where accurate coincidence detection directly impacts both the secret key rate and robustness against adversarial interference. Existing synchronization approaches typically assume benign timing environments, leaving open the possibility of subtle yet impactful adversarial attacks on the temporal domain. In this work, we introduce an adversary-aware synchronization framework for entanglement-assisted QKD that systematically models timing threats and proposes resilient countermeasures. First, we present a formal system model capturing the timing dependencies of entangled-photon-based QKD protocols. We then introduce a taxonomy of timing adversaries, including passive noise amplifiers, delay injectors, jitter inducers, and desynchronization attackers, each characterized by distinct operational footprints and security implications. To mitigate these threats, we propose a preliminary adaptive windowing algorithm that dynamically adjusts coincidence detection thresholds based on realtime statistical estimates of photon arrival distributions. Initial analysis and simulations demonstrate that the proposed adaptive strategy outperforms fixed-window synchronization under both natural noise and adversarial timing manipulations, achieving improved key rates while maintaining security guarantees. These results highlight the need for adversary-aware synchronization as a critical component of next-generation entanglement-assisted QKD systems.
Deep neural networks remain highly vulnerable to adversarial perturbations, limiting their reliability in security- and safety-critical applications. To address this challenge, we introduce QShield, a modular hybrid quantum-classical neural network (HQCNN) architecture designed to enhance the adversarial robustness of classical deep learning models. QShield integrates a conventional convolutional neural network (CNN) backbone for feature extraction with a quantum processing module that encodes the extracted features into quantum states, applies structured entanglement operations under realistic noise models, and outputs a hybrid prediction through a dynamically weighted fusion mechanism implemented via a lightweight multilayer perceptron (MLP). We systematically evaluate both classical and hybrid quantum-classical models on the MNIST, OrganAMNIST, and CIFAR-10 datasets, using a comprehensive set of robustness, efficiency, and computational performance metrics. Our results demonstrate that classical models are highly vulnerable to adversarial attacks, whereas the proposed hybrid models with entanglement patterns maintain high predictive accuracy while substantially reducing attack success rates across a wide range of adversarial attacks. Furthermore, the proposed hybrid architecture significantly increased the computational cost required to generate adversarial examples, thereby introducing an additional layer of defense. These findings indicate that the proposed modular hybrid architecture achieves a practical balance between predictive accuracy and adversarial robustness, positioning it as a promising approach for secure and reliable machine learning in sensitive and safety-critical applications.