The emergence of quantum computing poses significant risks to the security of modern communication networks as it breaks today's public-key cryptographic algorithms. Quantum Key Distribution (QKD) offers a promising solution by harnessing the principles of quantum mechanics to establish secure keys. However, practical QKD implementations remain vulnerable to hardware imperfections and advanced attacks such as Photon Number Splitting and Trojan-Horse attacks. In this work, we investigate the potential of quantum machine learning (QML) to detect QKD attacks. In particular, we propose a Hybrid Quantum Long Short-Term Memory (QLSTM) model to improve detection performance. By combining quantum-enhanced learning with classical deep learning, the model captures temporal patterns in QKD data, improving detection accuracy. To evaluate the proposed model, we introduce a QKD dataset that simulates typical operations along with multiple attack scenarios, including Intercept-and-Resend, Photon-Number Splitting, Trojan-Horse, Detector Blinding, and Combined attacks. The dataset includes Quantum Bit Error Rate (QBER), signal and decoy loss rates, and time-based metrics. Our results demonstrate the promising performance of the quantum machine learning approach compared to classical models. The proposed Hybrid QLSTM achieved an accuracy of 94.7% after 50 training epochs. The evaluation is conducted on a semi-realistic, simulation-generated decoy-state BB84 dataset, and the reported performance should be interpreted as a proof-of-concept rather than a final assessment on field-deployed QKD systems.

翻译：量子计算的出现对现代通信网络的安全性构成了重大威胁，因为它能够破解当今的公钥密码算法。量子密钥分发通过利用量子力学原理建立安全密钥，提供了一种有前景的解决方案。然而，实际的QKD实现仍易受硬件缺陷和高级攻击（如光子数分离攻击和特洛伊木马攻击）的影响。本研究探讨了量子机器学习在检测QKD攻击方面的潜力。具体而言，我们提出了一种混合量子长短期记忆模型以提高检测性能。该模型通过将量子增强学习与经典深度学习相结合，能够捕捉QKD数据中的时序模式，从而提升检测准确率。为评估所提出的模型，我们引入了一个QKD数据集，该数据集模拟了典型操作及多种攻击场景，包括拦截重发攻击、光子数分离攻击、特洛伊木马攻击、探测器致盲攻击及组合攻击。数据集包含量子比特错误率、信号与诱骗态损耗率以及基于时间的度量指标。我们的结果表明，与经典模型相比，量子机器学习方法展现出优越的性能。所提出的混合QLSTM模型在50个训练周期后达到了94.7%的准确率。评估基于半真实、仿真生成的诱骗态BB84协议数据集进行，所报告的性能应视为概念验证，而非对实际部署QKD系统的最终评估。