Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Applied Sciences

Adversarial Attack Defense Method for a Continuous-Variable Quantum Key Distribution System Based on Kernel Robust Manifold Non-Negative Matrix Factorization

Yuwen Fu, E Xia, Duan Huang, Yumei Jing

Published Sep 2, 2023• 13 citations
Publisher

Abstract

Machine learning has been applied in continuous-variable quantum key distribution (CVQKD) systems to address the growing threat of quantum hacking attacks. However, the use of machine learning algorithms for detecting these attacks has uncovered a vulnerability to adversarial disturbances that can compromise security. By subtly perturbing the detection networks used in CVQKD, significant misclassifications can occur. To address this issue, we utilize an adversarial sample defense method based on non-negative matrix factorization (NMF), considering the nonlinearity and high-dimensional nature of CVQKD data. Specifically, we employ the Kernel Robust Manifold Non-negative Matrix Factorization (KRMNMF) algorithm to reconstruct input samples, reducing the impact of adversarial perturbations. Firstly, we extract attack features against CVQKD by considering the adversary known as Eve. Then, we design an Artificial Neural Network (ANN) detection model to identify these attacks. Next, we introduce adversarial perturbations into the data generated by Eve. Finally, we use the KRMNMF decomposition to extract features from CVQKD data and mitigate the influence of adversarial perturbations through reconstruction. Experimental results demonstrate that the application of KRMNMF can effectively defend against adversarial attacks to a certain extent. The accuracy of KRMNMF surpasses the commonly used Comdefend method by 32.2% and the JPEG method by 30.8%. Moreover, it exhibits an improvement of 20.8% compared to NMF and outperforms other NMF-related algorithms in terms of classification accuracy. Moreover, it can complement other defense strategies, thus enhancing the overall defensive capabilities of CVQKD systems.

DOI
10.3390/app13179928

Related papers

Adversarial Perturbation Elimination with GAN Based Defense in Continuous-Variable Quantum Key Distribution Systems

Xun Tang, Pengzhi Yin, Zehao Zhou +1 more

Machine learning is being applied to continuous-variable quantum key distribution (CVQKD) systems as defense countermeasures for attack classification. However, recent studies have demonstrated that most of these detection networks are not immune to adversarial attacks. In this paper, we propose to implement typical adversarial attack strategies against the CVQKD system and introduce a generalized defense scheme. Adversarial attacks essentially generate data points located near decision boundaries that are linearized based on iterations of the classifier to lead to misclassification. Using the DeepFool attack as an example, we test it on four different CVQKD detection networks and demonstrate that an adversarial attack can fool most CVQKD detection networks. To solve this problem, we propose an improved adversarial perturbation elimination with a generative adversarial network (APE-GAN) scheme to generate samples with similar distribution to the original samples to defend against adversarial attacks. The results show that the proposed scheme can effectively defend against adversarial attacks including DeepFool and other adversarial attacks and significantly improve the security of communication systems.

Electronics•May 27, 2023•11 citations

Dictionary Learning Based Scheme for Adversarial Defense in Continuous-Variable Quantum Key Distribution

Shimiao Li, Pengzhi Yin, Zehao Zhou +3 more

There exist various attack strategies in continuous-variable quantum key distribution (CVQKD) system in practice. Due to the powerful information processing ability of neural networks, they are applied to the detection and classification of attack strategies in CVQKD systems. However, neural networks are vulnerable to adversarial attacks, resulting in the CVQKD system using neural networks also having security risks. To solve this problem, we propose a defense scheme for the CVQKD system. We first perform low-rank dimensionality reduction on the CVQKD system data through regularized self-representation-locality preserving projects (RSR-LPP) to filter out some adversarial disturbances, and then perform sparse coding reconstruction through dictionary learning to add data details and filter residual adversarial disturbances. We test the proposed defense algorithm in the CVQKD system. The results indicate that our proposed scheme has a good monitoring and alarm effect on CVQKD adversarial disturbances and has a better effect than other compared defense algorithms.

Entropy•Mar 14, 2023•5 citations

Adversarial Attacks Defense for Continuous Variable Quantum Key Distribution System Based on Diffusion Model

Keke Hu, Haiyang Hu, Li Ye +1 more

Deep learning is applied to continuous-variable quantum key distribution (CVQKD) systems for attack detection and classification. However, adversarial attacks can bypass these detection mechanisms, posing a significant threat. To address this, we introduce QuantumPure, a defense mechanism based on a diffusion model that purifies adversarial inputs. QuantumPure first introduces controlled noise to the adversarial data through a forward diffusion process, then applies a reverse sampling step to reconstruct clean data. This approach ensures that adversarial perturbations are neutralized while preserving the integrity of the quantum data. We demonstrate that QuantumPure enhances the robustness of CVQKD systems across a broad spectrum of adversarial attacks, encompassing both white-box and black-box scenarios while preserving high classification accuracy. Furthermore, the results demonstrate that our approach is resistant to unseen attack methods and greatly enhances the security of communication systems.

International Conference on Computing, Communication, Perception and Quantum Technology (CCPQT)•Oct 25, 2024•0 citations

Detecting quantum attacks: a machine learning based defense strategy for practical continuous-variable quantum key distribution

Yiyu Mao, Wenti Huang, Hai Zhong +4 more

Abstract The practical security of a continuous-variable quantum key distribution (CVQKD) system is compromised by various attack strategies. The existing countermeasures against these attacks are to exploit different real-time monitoring modules to prevent different types of attacks, which significantly depend on the accuracy of the estimated excess noise and lack a universal defense method. In this paper, we propose a defense strategy for CVQKD systems to address these disadvantages and resist most of the known attack types. We investigate several features of the pulses that would be affected by different types of attacks, derive a feature vector based on these features as the input of an artificial neural network (ANN) model, and show the training and testing process of the ANN model for attack detection and classification. Simulation results show that the proposed scheme can effectively detect most of the known attacks at the cost of reducing a small part of secret keys and transmission distance. It establishes a universal attack detection model by simply monitoring several features of the pulses without knowing the exact type of attack in advance.

New Journal of Physics•Jul 23, 2020•75 citations

Artificial key fingerprints for continuous-variable quantum key distribution

Yan Yao, Duan Huang, Pengzhi Yin +4 more

With rapid progress and significant successes in quantum hacking attacks, machine learning is being applied in continuous-variable quantum key distribution (CVQKD) systems. However, most machine-learning networks are shown to be vulnerable to adversarial examples. Thus, we propose a defense scheme with artificial key fingerprints for CVQKD systems. We first embed artificial key fingerprints into quadrature values, then exploit changes of fingerprints in the communication process to probe whether the samples have been manipulated. The paper verifies the effectiveness of our proposal based on four groups of samples that consist of normal measurement values and abnormal measurement values under four attacks. Simulation results show that the model can monitor sample changes in real time with high precision, and our scheme stays robust against adversarial examples attack. This proposal provides a different idea to address the potential vulnerability between machine-learning models used in CVQKD systems and their possible attacked training and test data by adversarial examples.

Physical Review A•Jul 5, 2023•3 citations

Machine-learning-based detection for quantum hacking attacks on continuous-variable quantum-key-distribution systems

Chao Ding, Shi Wang, Yaonan Wang +3 more

Continuous-variable quantum key distribution (CVQKD) is a mature technology that can theoretically provide an unconditional security guarantee. However, a practical CVQKD system may be vulnerable to various quantum hacking attacks due to imperfect devices and insufficient assumptions. In this paper, we propose a universal defense strategy called a machine-learning-based attack detection scheme (MADS). Leveraging the combined advantages of density-based spatial clustering of applications with noise (DBSCAN) and multiclass support vector machines (MCSVMs), MADS demonstrates remarkable effectiveness in detecting quantum hacking attacks. Specifically, we first establish a set of attack-related features to extract feature vectors. These vectors are then utilized as input data for DBSCAN to identify and remove any noise or outliers. Finally, the trained MCSVMs are employed to classify and predict the processed data. The predicted results can immediately determine whether or not to generate a final secret key. Simulation results show that the proposed MADS can efficiently detect most quantum hacking attacks and revise the overestimated secret key rates caused by a CVQKD system without any defense strategy to obtain a tighter security bound.

Physical Review A•Jun 28, 2023•14 citations
Machine learning has been applied in continuous-variable quantum key distribution (CVQKD) systems to address the growing threat of quantum hacking attacks. However, the use of machine learning algorithms for detecting these attacks has uncovered a vulnerability to adversarial disturbances that can compromise security. By subtly perturbing the detection networks used in CVQKD, significant misclassifications can occur. To address this issue, we utilize an adversarial sample defense method based on non-negative matrix factorization (NMF), considering the nonlinearity and high-dimensional nature of CVQKD data. Specifically, we employ the Kernel Robust Manifold Non-negative Matrix Factorization (KRMNMF) algorithm to reconstruct input samples, reducing the impact of adversarial perturbations. Firstly, we extract attack features against CVQKD by considering the adversary known as Eve. Then, we design an Artificial Neural Network (ANN) detection model to identify these attacks. Next, we introduce adversarial perturbations into the data generated by Eve. Finally, we use the KRMNMF decomposition to extract features from CVQKD data and mitigate the influence of adversarial perturbations through reconstruction. Experimental results demonstrate that the application of KRMNMF can effectively defend against adversarial attacks to a certain extent. The accuracy of KRMNMF surpasses the commonly used Comdefend method by 32.2% and the JPEG method by 30.8%. Moreover, it exhibits an improvement of 20.8% compared to NMF and outperforms other NMF-related algorithms in terms of classification accuracy. Moreover, it can complement other defense strategies, thus enhancing the overall defensive capabilities of CVQKD systems.
Machine learning is being applied to continuous-variable quantum key distribution (CVQKD) systems as defense countermeasures for attack classification. However, recent studies have demonstrated that most of these detection networks are not immune to adversarial attacks. In this paper, we propose to implement typical adversarial attack strategies against the CVQKD system and introduce a generalized defense scheme. Adversarial attacks essentially generate data points located near decision boundaries that are linearized based on iterations of the classifier to lead to misclassification. Using the DeepFool attack as an example, we test it on four different CVQKD detection networks and demonstrate that an adversarial attack can fool most CVQKD detection networks. To solve this problem, we propose an improved adversarial perturbation elimination with a generative adversarial network (APE-GAN) scheme to generate samples with similar distribution to the original samples to defend against adversarial attacks. The results show that the proposed scheme can effectively defend against adversarial attacks including DeepFool and other adversarial attacks and significantly improve the security of communication systems.
There exist various attack strategies in continuous-variable quantum key distribution (CVQKD) system in practice. Due to the powerful information processing ability of neural networks, they are applied to the detection and classification of attack strategies in CVQKD systems. However, neural networks are vulnerable to adversarial attacks, resulting in the CVQKD system using neural networks also having security risks. To solve this problem, we propose a defense scheme for the CVQKD system. We first perform low-rank dimensionality reduction on the CVQKD system data through regularized self-representation-locality preserving projects (RSR-LPP) to filter out some adversarial disturbances, and then perform sparse coding reconstruction through dictionary learning to add data details and filter residual adversarial disturbances. We test the proposed defense algorithm in the CVQKD system. The results indicate that our proposed scheme has a good monitoring and alarm effect on CVQKD adversarial disturbances and has a better effect than other compared defense algorithms.
Deep learning is applied to continuous-variable quantum key distribution (CVQKD) systems for attack detection and classification. However, adversarial attacks can bypass these detection mechanisms, posing a significant threat. To address this, we introduce QuantumPure, a defense mechanism based on a diffusion model that purifies adversarial inputs. QuantumPure first introduces controlled noise to the adversarial data through a forward diffusion process, then applies a reverse sampling step to reconstruct clean data. This approach ensures that adversarial perturbations are neutralized while preserving the integrity of the quantum data. We demonstrate that QuantumPure enhances the robustness of CVQKD systems across a broad spectrum of adversarial attacks, encompassing both white-box and black-box scenarios while preserving high classification accuracy. Furthermore, the results demonstrate that our approach is resistant to unseen attack methods and greatly enhances the security of communication systems.
Abstract The practical security of a continuous-variable quantum key distribution (CVQKD) system is compromised by various attack strategies. The existing countermeasures against these attacks are to exploit different real-time monitoring modules to prevent different types of attacks, which significantly depend on the accuracy of the estimated excess noise and lack a universal defense method. In this paper, we propose a defense strategy for CVQKD systems to address these disadvantages and resist most of the known attack types. We investigate several features of the pulses that would be affected by different types of attacks, derive a feature vector based on these features as the input of an artificial neural network (ANN) model, and show the training and testing process of the ANN model for attack detection and classification. Simulation results show that the proposed scheme can effectively detect most of the known attacks at the cost of reducing a small part of secret keys and transmission distance. It establishes a universal attack detection model by simply monitoring several features of the pulses without knowing the exact type of attack in advance.
With rapid progress and significant successes in quantum hacking attacks, machine learning is being applied in continuous-variable quantum key distribution (CVQKD) systems. However, most machine-learning networks are shown to be vulnerable to adversarial examples. Thus, we propose a defense scheme with artificial key fingerprints for CVQKD systems. We first embed artificial key fingerprints into quadrature values, then exploit changes of fingerprints in the communication process to probe whether the samples have been manipulated. The paper verifies the effectiveness of our proposal based on four groups of samples that consist of normal measurement values and abnormal measurement values under four attacks. Simulation results show that the model can monitor sample changes in real time with high precision, and our scheme stays robust against adversarial examples attack. This proposal provides a different idea to address the potential vulnerability between machine-learning models used in CVQKD systems and their possible attacked training and test data by adversarial examples.
Continuous-variable quantum key distribution (CVQKD) is a mature technology that can theoretically provide an unconditional security guarantee. However, a practical CVQKD system may be vulnerable to various quantum hacking attacks due to imperfect devices and insufficient assumptions. In this paper, we propose a universal defense strategy called a machine-learning-based attack detection scheme (MADS). Leveraging the combined advantages of density-based spatial clustering of applications with noise (DBSCAN) and multiclass support vector machines (MCSVMs), MADS demonstrates remarkable effectiveness in detecting quantum hacking attacks. Specifically, we first establish a set of attack-related features to extract feature vectors. These vectors are then utilized as input data for DBSCAN to identify and remove any noise or outliers. Finally, the trained MCSVMs are employed to classify and predict the processed data. The predicted results can immediately determine whether or not to generate a final secret key. Simulation results show that the proposed MADS can efficiently detect most quantum hacking attacks and revise the overestimated secret key rates caused by a CVQKD system without any defense strategy to obtain a tighter security bound.