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

IEEE photonics journal

Machine Learning Assisted Prediction for Free-Space Continuous Variable Quantum Teleportation

Jiaye Xu, Yun Jie Mao, Yuxin Chen, Xiaowen Xu, Ying Guo

Published Jun 27, 2022• 5 citations
Publisher

Abstract

Satellite-based quantum communication is critical in establishing a global quantum network for its ability to achieve thousand-kilometer-level tasks without the requirement of quantum relay. The free-space channel, however, is affected by diffraction, regular refraction, and atmospheric turbulence, which induce random wandering and broadening of the transmitted beam, resulting in the transmittance of fluctuation. In such a scenario, the ability to predict the unknown situation has become a problem worth considering. In this paper, we suggest a machine learning (ML) assisted prediction of continuous-variable quantum teleportation (CVQT) over satellite-ground channel. We apply the prediction algorithm, which involves$k$-nearest neighbor (KNN) algorithm and decision tree (DT) algorithm, to the squeezing parameter and the satellite altitude. Numerical analysis shows that the predicted parameter can reflect the actual derivation and calculation under certain circumstances within the allowable range of error. Moreover, we apply the prediction algorithm to the case affected by the fluctuating parameter transmittance. The results show that the ML-assisted prediction results are consistent with the actual data when zenith angles, turbulence strengths, and wavelengths are taken into account. It demonstrates the high identifying accuracy without increasing the complexity of the system.

DOI
10.1109/jphot.2022.3186391

Related papers

Overcoming the uplink limit of satellite-based quantum communication with deterministic quantum teleportation

Zhiyue Zuo, Yijun Wang, Qin Liao +1 more

Satellite-based quantum communication is essential for both foundational quantum physics tests and scalable quantum networks. However, the practical implementation of such application is limited due to the turbulent atmosphere, which is especially challenging for the ground-to-satellite uplink scenario. Here we consider the configuration of the continuous-variable quantum teleportation protocol to overcome both the uplink limit and background-radiation constraints with an unconditional and deterministic transfer, where the propagation of quantum light is affected by the favorable satellite-to-ground downlink (including the effects of absorption, scattering, turbulence, pointing errors, and background light). In particular, we derive the turbulent-induced fidelity that is achievable by the protocol for both nighttime and daytime operations, showing the effectiveness of this approach for various squeezing levels and orbit conditions (satellite altitudes and zenith angles). Finally, several approaches for fidelity enhancement are suggested, in terms of optimizing the transmitted Gaussian entanglement in the free-space scenario.

Physical Review A•Aug 31, 2021•23 citations

Machine Learning Assisted Continuous-Variable Quantum Communication

Nathan Long

This thesis investigates how machine learning (ML) can assist state-of-the-art continuous-variable quantum key distribution (CV-QKD) protocols to advance a global Quantum Internet. Relative wavefront errors between multiplexed classical reference pulses and quantum signals pose a significant challenge to the practical implementation of a satellite-Earth CV-QKD network across turbulent atmospheric channels, potentially leading to the destruction of secure key transfer. ML is proposed as a solution, where ML-based wavefront correction algorithms are designed to mitigate the potentially deleterious effects of relative wavefront errors on CV-QKD, as a result of excess noise affecting the quantum signals. This thesis provides the first analysis of two-dimensional spatial relative wavefront errors. The potential causes of relative wavefront errors are explored, their evolution across satellite-Earth channels is simulated, and their effect on achievable secure key rates is evaluated. ML-based wavefront correction algorithms are designed to facilitate enhanced key rates by encoding the wavefront corrections onto a real local oscillator at the receiver. The analysis of relative wavefront errors presented in this thesis culminates in a series of experimental campaigns involving the transmission of multiplexed stronger reference signals and weaker signals across a 2.4~km free-space optical link, providing the first experimental evidence of relative wavefront errors in the context of CV-QKD. Modified ML-based wavefront correction algorithms are then applied to the experimental data, using phase retrieval, resulting in up to a 2/3 reduction in relative phase error variance. Overall, it is shown that ML algorithms can augment state-of-the-art CV-QKD protocols across turbulent atmospheric channels, thereby progressing the practical feasibility of developing a truly global CV-QKD network.

•Jun 5, 2026•0 citations

Distillation-Enhanced Continuous-Variable Quantum Teleportation for Satellite Communication Networks

Lia Suci Waliani, Georges Kaddoum, Mahdi Chehimi +1 more

Quantum teleportation (QT) over satellite-based free-space optical (FSO) channels is a promising approach for long-distance quantum communication. However, its performance is significantly degraded by atmospheric loss and turbulence. In this paper, we investigate continuous-variable (CV) QT in a dual-downlink scenario, where a satellite distributes entangled states to two ground stations. To mitigate channel-induced degradation, we employ a non-Gaussian entanglement distillation protocol based on the sequential application of photon addition and photon subtraction (PA-PS) on the weaker channel. The results show that the proposed scheme improves teleportation fidelity by up to 7.7% and enhances entanglement negativity by approximately 105% in the low-to-moderate (below 600 km) loss regime. In addition, we identify an optimal squeezing parameter that balances entanglement strength and noise sensitivity. Taken together, these results demonstrate the effectiveness of PA-PS distillation for improving CV quantum communication in realistic satellite networks. We further characterize the trade-off between fidelity gain and the heralded success probability of the protocol.

quant-pharXiv:2607.08977•Jul 13, 2026•0 citations

Satellite-based entanglement distribution and quantum teleportation with continuous variables

Tasio Gonzalez-Raya, Stefano Pirandola, Mikel Sanz

Abstract Advances in satellite quantum communications aim at reshaping the global telecommunication network by increasing the security of the transferred information. Here, we study the effects of atmospheric turbulence in continuous-variable entanglement distribution and quantum teleportation in the optical regime between a ground station and a satellite. More specifically, we study the degradation of entanglement due to various error sources in the distribution, namely, diffraction, atmospheric attenuation, turbulence, and detector inefficiency, in both downlink and uplink scenarios. As the fidelity of a quantum teleportation protocol using these distributed entangled resources is not sufficient, we include an intermediate station for either state generation, or beam refocusing, in order to reduce the effects of atmospheric turbulence and diffraction, respectively. The results show the feasibility of free-space entanglement distribution and quantum teleportation in downlink paths up to the LEO region, but also in uplink paths with the help of the intermediate station. Finally, we complete the study with microwave-optical comparison in bad weather situations, and with the study of horizontal paths in ground-to-ground and inter-satellite quantum communication.

arXiv:2303.17224•Communications Physics•Apr 11, 2024•68 citations

Lightweight wavefront prediction for atmospheric turbulence in free-space CVQKD systems

Tian Zheng, Lanjian Chen, bozhi wei +3 more

Satellite-to-ground free-space optical links enable space-based quantum key distribution (QKD) Satellite-to-ground free-space optical links provide an important platform for space-based quantum key distribution (QKD) networks. Continuous-variable quantum key distribution (CVQKD) is compatible with coherent optical communication systems, making it a promising approach for satellite-to-ground QKD. However, atmospheric turbulence introduces wavefront distortion and intensity fluctuations that degrade optical coupling and limit CVQKD performance. Conventional adaptive optics (AO) compensation may involve substantial computational complexity and latency, motivating efficient wavefront prediction for rapidly varying atmospheric channels. In this work, we propose a lightweight Att-ConvGRU-UNet framework for predictive wavefront compensation. The framework combines a multi-scale UNet for spatial feature extraction, a ConvGRU module for temporal modeling, and a convolutional block attention module (CBAM) for adaptive channel- and spatial-feature refinement. Residual learning is employed to predict phase increments and improve training stability. In addition to conventional wavefront metrics, single-mode fiber (SMF) mixing efficiency and secret key rate (SKR) are used as downstream metrics to evaluate the impact of prediction performance on CVQKD. The proposed model achieves a mean squared error (MSE) of 0.0030 rad 2 and a structural similarity index (SSIM) of 0.9865 on a three-layer dynamic non-Kolmogorov phase-screen dataset, with 0.342 M trainable parameters. Predictor-side latency is further evaluated at different direct prediction horizons to characterize the trade-off between prediction accuracy and computational time. This work provides a lightweight and computationally efficient predictive solution for atmospheric turbulence compensation in satellite-to-ground CVQKD, offering a promising approach for next-generation space-based quantum communication networks.

EPJ Quantum Technology•Nov 13, 2026•0 citations

Passive Continuous Variable Measurement-Device-Independent Quantum Key Distribution Predictable with Machine Learning in Oceanic Turbulence

Jianmin Yi, Hao Wu, Ying Guo

Building an underwater quantum network is necessary for various applications such as ocean exploration, environmental monitoring, and national defense. Motivated by characteristics of the oceanic turbulence channel, we suggest a machine learning approach to predicting the channel characteristics of continuous variable (CV) quantum key distribution (QKD) in challenging seawater environments. We consider the passive continuous variable (CV) measurement-device-independent (MDI) QKD in oceanic scenarios, since the passive-state preparation scheme offers simpler linear elements for preparation, resulting in reduced interaction with the practical environment. To provide a practical reference for underwater quantum communications, we suggest a prediction of transmittance for the ocean quantum links with a given neural network as an example of machine learning algorithms. The results have a good consistency with the real data within the allowable error range; this makes the passive CVQKD more promising for commercialization and implementation.

Entropy•Feb 27, 2024•3 citations
Satellite-based quantum communication is critical in establishing a global quantum network for its ability to achieve thousand-kilometer-level tasks without the requirement of quantum relay. The free-space channel, however, is affected by diffraction, regular refraction, and atmospheric turbulence, which induce random wandering and broadening of the transmitted beam, resulting in the transmittance of fluctuation. In such a scenario, the ability to predict the unknown situation has become a problem worth considering. In this paper, we suggest a machine learning (ML) assisted prediction of continuous-variable quantum teleportation (CVQT) over satellite-ground channel. We apply the prediction algorithm, which involveskkk-nearest neighbor (KNN) algorithm and decision tree (DT) algorithm, to the squeezing parameter and the satellite altitude. Numerical analysis shows that the predicted parameter can reflect the actual derivation and calculation under certain circumstances within the allowable range of error. Moreover, we apply the prediction algorithm to the case affected by the fluctuating parameter transmittance. The results show that the ML-assisted prediction results are consistent with the actual data when zenith angles, turbulence strengths, and wavelengths are taken into account. It demonstrates the high identifying accuracy without increasing the complexity of the system.
Satellite-based quantum communication is essential for both foundational quantum physics tests and scalable quantum networks. However, the practical implementation of such application is limited due to the turbulent atmosphere, which is especially challenging for the ground-to-satellite uplink scenario. Here we consider the configuration of the continuous-variable quantum teleportation protocol to overcome both the uplink limit and background-radiation constraints with an unconditional and deterministic transfer, where the propagation of quantum light is affected by the favorable satellite-to-ground downlink (including the effects of absorption, scattering, turbulence, pointing errors, and background light). In particular, we derive the turbulent-induced fidelity that is achievable by the protocol for both nighttime and daytime operations, showing the effectiveness of this approach for various squeezing levels and orbit conditions (satellite altitudes and zenith angles). Finally, several approaches for fidelity enhancement are suggested, in terms of optimizing the transmitted Gaussian entanglement in the free-space scenario.
This thesis investigates how machine learning (ML) can assist state-of-the-art continuous-variable quantum key distribution (CV-QKD) protocols to advance a global Quantum Internet. Relative wavefront errors between multiplexed classical reference pulses and quantum signals pose a significant challenge to the practical implementation of a satellite-Earth CV-QKD network across turbulent atmospheric channels, potentially leading to the destruction of secure key transfer. ML is proposed as a solution, where ML-based wavefront correction algorithms are designed to mitigate the potentially deleterious effects of relative wavefront errors on CV-QKD, as a result of excess noise affecting the quantum signals. This thesis provides the first analysis of two-dimensional spatial relative wavefront errors. The potential causes of relative wavefront errors are explored, their evolution across satellite-Earth channels is simulated, and their effect on achievable secure key rates is evaluated. ML-based wavefront correction algorithms are designed to facilitate enhanced key rates by encoding the wavefront corrections onto a real local oscillator at the receiver. The analysis of relative wavefront errors presented in this thesis culminates in a series of experimental campaigns involving the transmission of multiplexed stronger reference signals and weaker signals across a 2.4~km free-space optical link, providing the first experimental evidence of relative wavefront errors in the context of CV-QKD. Modified ML-based wavefront correction algorithms are then applied to the experimental data, using phase retrieval, resulting in up to a 2/3 reduction in relative phase error variance. Overall, it is shown that ML algorithms can augment state-of-the-art CV-QKD protocols across turbulent atmospheric channels, thereby progressing the practical feasibility of developing a truly global CV-QKD network.
Quantum teleportation (QT) over satellite-based free-space optical (FSO) channels is a promising approach for long-distance quantum communication. However, its performance is significantly degraded by atmospheric loss and turbulence. In this paper, we investigate continuous-variable (CV) QT in a dual-downlink scenario, where a satellite distributes entangled states to two ground stations. To mitigate channel-induced degradation, we employ a non-Gaussian entanglement distillation protocol based on the sequential application of photon addition and photon subtraction (PA-PS) on the weaker channel. The results show that the proposed scheme improves teleportation fidelity by up to 7.7% and enhances entanglement negativity by approximately 105% in the low-to-moderate (below 600 km) loss regime. In addition, we identify an optimal squeezing parameter that balances entanglement strength and noise sensitivity. Taken together, these results demonstrate the effectiveness of PA-PS distillation for improving CV quantum communication in realistic satellite networks. We further characterize the trade-off between fidelity gain and the heralded success probability of the protocol.
Abstract Advances in satellite quantum communications aim at reshaping the global telecommunication network by increasing the security of the transferred information. Here, we study the effects of atmospheric turbulence in continuous-variable entanglement distribution and quantum teleportation in the optical regime between a ground station and a satellite. More specifically, we study the degradation of entanglement due to various error sources in the distribution, namely, diffraction, atmospheric attenuation, turbulence, and detector inefficiency, in both downlink and uplink scenarios. As the fidelity of a quantum teleportation protocol using these distributed entangled resources is not sufficient, we include an intermediate station for either state generation, or beam refocusing, in order to reduce the effects of atmospheric turbulence and diffraction, respectively. The results show the feasibility of free-space entanglement distribution and quantum teleportation in downlink paths up to the LEO region, but also in uplink paths with the help of the intermediate station. Finally, we complete the study with microwave-optical comparison in bad weather situations, and with the study of horizontal paths in ground-to-ground and inter-satellite quantum communication.
Satellite-to-ground free-space optical links enable space-based quantum key distribution (QKD) Satellite-to-ground free-space optical links provide an important platform for space-based quantum key distribution (QKD) networks. Continuous-variable quantum key distribution (CVQKD) is compatible with coherent optical communication systems, making it a promising approach for satellite-to-ground QKD. However, atmospheric turbulence introduces wavefront distortion and intensity fluctuations that degrade optical coupling and limit CVQKD performance. Conventional adaptive optics (AO) compensation may involve substantial computational complexity and latency, motivating efficient wavefront prediction for rapidly varying atmospheric channels. In this work, we propose a lightweight Att-ConvGRU-UNet framework for predictive wavefront compensation. The framework combines a multi-scale UNet for spatial feature extraction, a ConvGRU module for temporal modeling, and a convolutional block attention module (CBAM) for adaptive channel- and spatial-feature refinement. Residual learning is employed to predict phase increments and improve training stability. In addition to conventional wavefront metrics, single-mode fiber (SMF) mixing efficiency and secret key rate (SKR) are used as downstream metrics to evaluate the impact of prediction performance on CVQKD. The proposed model achieves a mean squared error (MSE) of 0.0030 rad 2 and a structural similarity index (SSIM) of 0.9865 on a three-layer dynamic non-Kolmogorov phase-screen dataset, with 0.342 M trainable parameters. Predictor-side latency is further evaluated at different direct prediction horizons to characterize the trade-off between prediction accuracy and computational time. This work provides a lightweight and computationally efficient predictive solution for atmospheric turbulence compensation in satellite-to-ground CVQKD, offering a promising approach for next-generation space-based quantum communication networks.
Building an underwater quantum network is necessary for various applications such as ocean exploration, environmental monitoring, and national defense. Motivated by characteristics of the oceanic turbulence channel, we suggest a machine learning approach to predicting the channel characteristics of continuous variable (CV) quantum key distribution (QKD) in challenging seawater environments. We consider the passive continuous variable (CV) measurement-device-independent (MDI) QKD in oceanic scenarios, since the passive-state preparation scheme offers simpler linear elements for preparation, resulting in reduced interaction with the practical environment. To provide a practical reference for underwater quantum communications, we suggest a prediction of transmittance for the ocean quantum links with a given neural network as an example of machine learning algorithms. The results have a good consistency with the real data within the allowable error range; this makes the passive CVQKD more promising for commercialization and implementation.