Hongyu Pei
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hongyu Pei has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
- Wei Quan
School of Instrumentation and Optoelectronic Engineering, Beihang University 1 , Beijing 100191
7 shared publications - Wenfeng Fan
School of Instrumentation and Optoelectronic Engineering, Beihang University 1 , Beijing 100191
7 shared publications - Lihong Duan
School of Instrumentation and Optoelectronic Engineering, Beihang University, Beijing 100191, People’s Republic of China
3 shared publications Wei Quan
3 shared publicationsWenfeng Fan
3 shared publications- Haoying Pang
Beihang University
2 shared publications
Latest Publications
- 20 April 2026
- Published by AIP Publishing in Applied Physics Letters
- Vol. 128 (16)
Decoupling and suppression of opto-thermal drifts in SERF atomic co-magnetometers via precise null-point characterization
The long-term bias stability of spin-exchange relaxation-free atomic co-magnetometers is significantly constrained by coupled opto-thermal fluctuations. Traditionally, random systemic offsets at the magnetic compensation point lead to inconsistent opto-thermal coupling, compromising the reproducibility of sensor performance. In this work, we propose a method to simultaneously decouple and suppress these primary drifts by precisely characterizing the physical null-point of the atomic ensemble. A combined compensation scheme, joining residual rotation angle alignment and transverse magnetic field adjustment, is implemented to isolate the sensor output from fluctuations in probe laser intensity and vapor cell temperature. Experimental results demonstrate that the sensitivities to temperature drifts and probe power fluctuations are suppressed simultaneously at the identified null-point by factors of 72 and 880, respectively. This approach establishes a stable zero point, which is expected to significantly enhance sensor reproducibility and stability without sacrificing fundamental sensing performance.
- 20 April 2026
- Published by AIP Publishing in Applied Physics Letters
- Vol. 128 (16)
Decoupling and suppression of opto-thermal drifts in SERF atomic co-magnetometers via precise null-point characterization
The long-term bias stability of spin-exchange relaxation-free atomic co-magnetometers is significantly constrained by coupled opto-thermal fluctuations. Traditionally, random systemic offsets at the magnetic compensation point lead to inconsistent opto-thermal coupling, compromising the reproducibility of sensor performance. In this work, we propose a method to simultaneously decouple and suppress these primary drifts by precisely characterizing the physical null-point of the atomic ensemble. A combined compensation scheme, joining residual rotation angle alignment and transverse magnetic field adjustment, is implemented to isolate the sensor output from fluctuations in probe laser intensity and vapor cell temperature. Experimental results demonstrate that the sensitivities to temperature drifts and probe power fluctuations are suppressed simultaneously at the identified null-point by factors of 72 and 880, respectively. This approach establishes a stable zero point, which is expected to significantly enhance sensor reproducibility and stability without sacrificing fundamental sensing performance.
- 22 October 2025
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 113
Deep learning with explainability: Improving diagnostic accuracy and interpretability in ECG and MCG analysis
- 22 October 2025
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 113
Deep learning with explainability: Improving diagnostic accuracy and interpretability in ECG and MCG analysis
- 29 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-10
In Situ Synchronous Measurement of Low-Frequency Magnetic Field Fluctuation Based on Temporal Differencing in Atomic Comagnetometer
Fluctuations in transverse low-frequency magnetic fields significantly impact the long-term stability of atomic comagnetometers (ACMs). Due to the strong coupling between magnetic fields and gyro signals, conventional evaluation approaches are typically limited to offline analysis. To address this challenge, this paper proposes a temporal differencing (TD) technique combined with baseline alignment (BA), leveraging quantum spin dynamics under longitudinal magnetic field modulation to construct a differential response model. The TD approach compresses the response time scale of the system, thereby enhancing immunity to multi-source disturbances. Meanwhile, BA enables the synchronous acquisition of magnetic field fluctuations and gyro signals, facilitating real-time evaluation. Experimental results show that, under disturbances such as probe light power, pump light power, and vapor cell temperature, the proposed TD method achieves a suppression of at least 56.5% compared to INSR and a suppression of at least 65.6% compared to the Z-mode method, covering all components of the transverse field. Furthermore, the equivalent angular velocity deviations caused by transverse field fluctuations are quantified as 6.9×10-3 deg/h along the x-axis and 2.9×10-3 deg/h along the y-axis, contributing 28.4% and 11.9% to the degradation of the long-term stability of the ACM, respectively. These results demonstrate that the proposed method enables accurate and reliable in-situ evaluation of transverse magnetic field disturbances and offers practical insights for enhancing the robustness of ACM systems.
- 29 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-10
In Situ Synchronous Measurement of Low-Frequency Magnetic Field Fluctuation Based on Temporal Differencing in Atomic Comagnetometer
Fluctuations in transverse low-frequency magnetic fields significantly impact the long-term stability of atomic comagnetometers (ACMs). Due to the strong coupling between magnetic fields and gyro signals, conventional evaluation approaches are typically limited to offline analysis. To address this challenge, this paper proposes a temporal differencing (TD) technique combined with baseline alignment (BA), leveraging quantum spin dynamics under longitudinal magnetic field modulation to construct a differential response model. The TD approach compresses the response time scale of the system, thereby enhancing immunity to multi-source disturbances. Meanwhile, BA enables the synchronous acquisition of magnetic field fluctuations and gyro signals, facilitating real-time evaluation. Experimental results show that, under disturbances such as probe light power, pump light power, and vapor cell temperature, the proposed TD method achieves a suppression of at least 56.5% compared to INSR and a suppression of at least 65.6% compared to the Z-mode method, covering all components of the transverse field. Furthermore, the equivalent angular velocity deviations caused by transverse field fluctuations are quantified as 6.9×10-3 deg/h along the x-axis and 2.9×10-3 deg/h along the y-axis, contributing 28.4% and 11.9% to the degradation of the long-term stability of the ACM, respectively. These results demonstrate that the proposed method enables accurate and reliable in-situ evaluation of transverse magnetic field disturbances and offers practical insights for enhancing the robustness of ACM systems.
- 19 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-12
Temporal Decomposition and Attention-Based Deep Learning for Magnetocardiography Signal Denoising
This paper proposes Temporal-decomposition and Attention-based Denoising Network (TADNet) for magnetocardiography (MCG) signal denoising. The raw time-series data are decomposed into trend and seasonal components. The trend component is modeled using a multi-layer perceptron to capture the trend of the data. For seasonal component, a multi-layer convolutional neural network is employed to perform segmentation and extract local temporal features. These fine-grained features are processed by a Transformer-based self-attention mechanism to capture long-range temporal dependencies. Finally, deconvolution operations are applied to reconstruct the signal, achieving effective noise reduction for MCG data. This paper conducts in-depth studies on effects of time, wavelet, and Fourier style attention on model performance, enhancing the interpretability of the model. Extensive experiments on clinical MCG data under various noise conditions demonstrate that TADNet outperforms traditional algorithms, proving its effectiveness and robustness. The research results indicate that TADNet provides a highly efficient and accurate solution for MCG signal denoising.
- 19 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-12
Temporal Decomposition and Attention-Based Deep Learning for Magnetocardiography Signal Denoising
This paper proposes Temporal-decomposition and Attention-based Denoising Network (TADNet) for magnetocardiography (MCG) signal denoising. The raw time-series data are decomposed into trend and seasonal components. The trend component is modeled using a multi-layer perceptron to capture the trend of the data. For seasonal component, a multi-layer convolutional neural network is employed to perform segmentation and extract local temporal features. These fine-grained features are processed by a Transformer-based self-attention mechanism to capture long-range temporal dependencies. Finally, deconvolution operations are applied to reconstruct the signal, achieving effective noise reduction for MCG data. This paper conducts in-depth studies on effects of time, wavelet, and Fourier style attention on model performance, enhancing the interpretability of the model. Extensive experiments on clinical MCG data under various noise conditions demonstrate that TADNet outperforms traditional algorithms, proving its effectiveness and robustness. The research results indicate that TADNet provides a highly efficient and accurate solution for MCG signal denoising.
- 17 March 2025
- Published by Wiley in Advanced Quantum Technologies
- Vol. 8 (7)
Zero‐Position Closed‐Loop Control of Nuclear Spin in a K‐Rb‐21Ne$^{21}{\rm Ne}$ Comagnetometer
In SERF (spin‐exchange relaxation‐free) comagnetometers, the fluctuation of the atomic ensemble's polarization significantly impacts the long‐term stability of the output signal. A real‐time closed‐loop control method for the nuclear spin polarization is proposed, and the measurement model is established using Kalman state observer. Furthermore, magnetic field control signals are generated using a proportional‐integral‐derivative (PID) strategy, which is then applied to the atomic ensemble to dynamically adjust the nuclear spin polarization. Finally, closed‐loop control of nuclear spin polarization is achieved through continuous feedback and control. Moreover, experiments have demonstrated that implementing a closed‐loop system for nuclear spin zero‐position is conducive to reducing the sensitivity of the output signal to fluctuations in the external environment. The experimental results show that the Allan deviation at 100 s is reduced by , and the sensitivity of inertial rotation measurement is significantly improved, that is, the noise level at 1 Hz is suppressed by approximately .
- 17 March 2025
- Published by Wiley in Advanced Quantum Technologies
- Vol. 8 (7)
Zero‐Position Closed‐Loop Control of Nuclear Spin in a K‐Rb‐21Ne$^{21}{\rm Ne}$ Comagnetometer
In SERF (spin‐exchange relaxation‐free) comagnetometers, the fluctuation of the atomic ensemble's polarization significantly impacts the long‐term stability of the output signal. A real‐time closed‐loop control method for the nuclear spin polarization is proposed, and the measurement model is established using Kalman state observer. Furthermore, magnetic field control signals are generated using a proportional‐integral‐derivative (PID) strategy, which is then applied to the atomic ensemble to dynamically adjust the nuclear spin polarization. Finally, closed‐loop control of nuclear spin polarization is achieved through continuous feedback and control. Moreover, experiments have demonstrated that implementing a closed‐loop system for nuclear spin zero‐position is conducive to reducing the sensitivity of the output signal to fluctuations in the external environment. The experimental results show that the Allan deviation at 100 s is reduced by , and the sensitivity of inertial rotation measurement is significantly improved, that is, the noise level at 1 Hz is suppressed by approximately .
- 6 November 2024
- Published by Elsevier in Measurement
- Vol. 242
Bandwidth compensation for ultra-high-sensitivity SERF magnetometers in magnetocardiac sensing
- 6 November 2024
- Published by Elsevier in Measurement
- Vol. 242
Bandwidth compensation for ultra-high-sensitivity SERF magnetometers in magnetocardiac sensing
- 4 April 2024
- Published by Elsevier in Measurement
- Vol. 231
Pulsed optical pumping in electron spin vapor
- 4 April 2024
- Published by Elsevier in Measurement
- Vol. 231
Pulsed optical pumping in electron spin vapor
- Open Access
- 13 February 2024
- Published by Elsevier in Cell Reports Physical Science
- Vol. 5 (2)
Real-time quantum control of spin-coupling damping and application in atomic spin gyroscopes
- 13 February 2024
- Published by Elsevier in Cell Reports Physical Science
- Vol. 5 (2)
Open AccessReal-time quantum control of spin-coupling damping and application in atomic spin gyroscopes
- 18 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-9
Real-Time In Situ Quantum Feedback Control of Electron Spin in Atomic Spin Gyroscopes
The atomic spin gyroscope (ASG), based on spin-exchange relaxation-free (SERF) co-magnetometers, is poised to become the next generation of high-precision inertial sensors. However, the long-term stability of the gyro can be limited by electron spin polarization. In this study, we investigate a novel real-time, in situ quantum feedback control technique for electron spin polarization in ASGs. By varying the pumping rates, we analyze the gyroscope’s response to the modulation magnetic field, finding that electron polarization dramatically affects the magnetic field response properties of ASGs. It’s demonstrated that the amplitude or phase of the electron spin resonance can be locked for in situ control of electron spin polarization in real-time. A K-Rb-21Ne co-magnetometer is established to implement the proposed method, where a transverse modulating magnetic field is applied to resonate with the electron spin. The lock-in amplification (LIA) technique is used to obtain the amplitude or phase of the electron spin resonance, and the electron spin polarization is closed-loop controlled by locking the second harmonic resonance amplitude or phase. The gyro signal is extracted from the first harmonic of the electron spin resonance. Our experimental results show that real-time in situ feedback control of electron spin significantly improves ASG’s long-term stability and suppresses low-frequency noise. This proposed method has broad application prospects, including optically pumped magnetometers (OPMs), nuclear magnetic resonance gyroscopes (NMRGs), and co-magnetometers.
- 18 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-9
Real-Time In Situ Quantum Feedback Control of Electron Spin in Atomic Spin Gyroscopes
The atomic spin gyroscope (ASG), based on spin-exchange relaxation-free (SERF) co-magnetometers, is poised to become the next generation of high-precision inertial sensors. However, the long-term stability of the gyro can be limited by electron spin polarization. In this study, we investigate a novel real-time, in situ quantum feedback control technique for electron spin polarization in ASGs. By varying the pumping rates, we analyze the gyroscope’s response to the modulation magnetic field, finding that electron polarization dramatically affects the magnetic field response properties of ASGs. It’s demonstrated that the amplitude or phase of the electron spin resonance can be locked for in situ control of electron spin polarization in real-time. A K-Rb-21Ne co-magnetometer is established to implement the proposed method, where a transverse modulating magnetic field is applied to resonate with the electron spin. The lock-in amplification (LIA) technique is used to obtain the amplitude or phase of the electron spin resonance, and the electron spin polarization is closed-loop controlled by locking the second harmonic resonance amplitude or phase. The gyro signal is extracted from the first harmonic of the electron spin resonance. Our experimental results show that real-time in situ feedback control of electron spin significantly improves ASG’s long-term stability and suppresses low-frequency noise. This proposed method has broad application prospects, including optically pumped magnetometers (OPMs), nuclear magnetic resonance gyroscopes (NMRGs), and co-magnetometers.
- 12 July 2023
- Published by Wiley in Advanced Quantum Technologies
- Vol. 6 (9)
Spin‐Decoupling and In‐Situ Nuclear Magnetometer with Hyperpolarized Nuclear Spin
This paper investigates the dynamical properties of the spin‐coupling ensemble, with a focus on the spin‐decoupling phenomenon. The nuclear spin in such systems is significantly impacted by the electron spin when the main magnetic field approaches the Fermi contact field of nuclear spin, leading to measurement inaccuracies. A comprehensive dynamics model of the electron‐nuclear spin coupling ensemble has been established and analyzed. It proposes that the primary magnetic field augmenting effectively disentangles the precession frequency of the two spin‐coupling species, thereby yielding a spin‐decoupled state. Such a state allows nuclear spin serves as a high sensitive element for magnetic field measurements, while electron spin serves as a probe to acquire the precession of nuclear spin. With proposed method, a magnetic field resolution of 0.75 nT in a 12000 nT environment is achieved, indicating potential for magnetic field detection in geomagnetic environments. It further demonstrates measurements of electron spin Fermi contact interaction and polarization based on spin‐decoupling and in situ nuclear spin precession, results indicate measurement errors close to 1%. These results demonstrate the effectiveness of our approach in improving the accuracy and sensitivity of magnetic field measurements in spin‐coupled systems such as co‐magnetometers, nuclear magnetic resonance gyroscopes, and nuclear magnetometers.
- 12 July 2023
- Published by Wiley in Advanced Quantum Technologies
- Vol. 6 (9)
Spin‐Decoupling and In‐Situ Nuclear Magnetometer with Hyperpolarized Nuclear Spin
This paper investigates the dynamical properties of the spin‐coupling ensemble, with a focus on the spin‐decoupling phenomenon. The nuclear spin in such systems is significantly impacted by the electron spin when the main magnetic field approaches the Fermi contact field of nuclear spin, leading to measurement inaccuracies. A comprehensive dynamics model of the electron‐nuclear spin coupling ensemble has been established and analyzed. It proposes that the primary magnetic field augmenting effectively disentangles the precession frequency of the two spin‐coupling species, thereby yielding a spin‐decoupled state. Such a state allows nuclear spin serves as a high sensitive element for magnetic field measurements, while electron spin serves as a probe to acquire the precession of nuclear spin. With proposed method, a magnetic field resolution of 0.75 nT in a 12000 nT environment is achieved, indicating potential for magnetic field detection in geomagnetic environments. It further demonstrates measurements of electron spin Fermi contact interaction and polarization based on spin‐decoupling and in situ nuclear spin precession, results indicate measurement errors close to 1%. These results demonstrate the effectiveness of our approach in improving the accuracy and sensitivity of magnetic field measurements in spin‐coupled systems such as co‐magnetometers, nuclear magnetic resonance gyroscopes, and nuclear magnetometers.