Yinfeng Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yinfeng Wu has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Renjian Feng
Key Laboratory of Education Ministry for Precision Opto-Mechatronics Technology, Beihang University, China
19 shared publications - Ning Yu
Key Laboratory of Education Ministry for Precision Opto-Mechatronics Technology, Beihang University, China
16 shared publications - Jiaxuan Yan
Key Laboratory of Education Ministry for Precision Opto-Mechatronics Technology and the School of Instrumentation and Optoelectronics Engineering, Beihang University, Beijing, China
4 shared publications - Xiaofeng Ma
Department of Electrical Engineering and Automation, Aalto University, Espoo, Finland
3 shared publications - Jiangwen Wan
School of Instrumentation Science and Opto-electronics Engineering, Beihang University, Beijing 100191, China
2 shared publications - Jiangwen Wan
School of Instrument Science and Opto-electronics Engineering, Beijing University of Aeronautics and Astronautics (Beihang University), Beijing 100191, China
2 shared publications
Latest Publications
- Open Access
- 20 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 77356-77366
Optimal Temporal SNR Window for Multi-Sensor Fusion: Theory and Jamming-Aware GNN Architecture for Electronic Warfare
Multi-object tracking in electronic warfare (EW) relies heavily on temporal fusion—accumulating observations across frames through filtering or recurrent models. Yet this strategy relies on the premise that inter-frame data association is sufficiently accurate to benefit from accumulation. We challenge this assumption by analyzing the temporal signal-to-noise ratio SNRt = Δp/σ (inter-frame displacement over measurement error) and deriving an Optimal SNR Window Theorem. Under nearest-neighbor association, temporal modeling is harmful when SNRt ≲ 5%, beneficial between ≈5% and ≈45%, and saturated beyond; these boundaries are operating-region markers under the assumed association model. In EW scenarios with heavy jamming and false-target injection, SNRt can fall to ∼2.3%—well inside the harmful region.We therefore design a single-frame graph attention network (GAT) with jamming-aware gating that bypasses temporal association entirely. On a multi-sensor benchmark (900 frames, 37.5% jammer ratio, 1 200 injected false targets), the proposed method achieves OSPA of 2717±449 and MOTA of 14.0%±9.3% across seven seeds, outperforming both classical filters (JPDA, LMB/GLMB) and all temporal GNN variants, which produce negative MOTA. A sweep across six noise levels confirms the predicted non-monotonic benefit curve, peaking at SNRt = 11.5% with negative gain at the 2.3% operating point.
- 20 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 77356-77366
Open AccessOptimal Temporal SNR Window for Multi-Sensor Fusion: Theory and Jamming-Aware GNN Architecture for Electronic Warfare
Multi-object tracking in electronic warfare (EW) relies heavily on temporal fusion—accumulating observations across frames through filtering or recurrent models. Yet this strategy relies on the premise that inter-frame data association is sufficiently accurate to benefit from accumulation. We challenge this assumption by analyzing the temporal signal-to-noise ratio SNRt = Δp/σ (inter-frame displacement over measurement error) and deriving an Optimal SNR Window Theorem. Under nearest-neighbor association, temporal modeling is harmful when SNRt ≲ 5%, beneficial between ≈5% and ≈45%, and saturated beyond; these boundaries are operating-region markers under the assumed association model. In EW scenarios with heavy jamming and false-target injection, SNRt can fall to ∼2.3%—well inside the harmful region.We therefore design a single-frame graph attention network (GAT) with jamming-aware gating that bypasses temporal association entirely. On a multi-sensor benchmark (900 frames, 37.5% jammer ratio, 1 200 injected false targets), the proposed method achieves OSPA of 2717±449 and MOTA of 14.0%±9.3% across seven seeds, outperforming both classical filters (JPDA, LMB/GLMB) and all temporal GNN variants, which produce negative MOTA. A sweep across six noise levels confirms the predicted non-monotonic benefit curve, peaking at SNRt = 11.5% with negative gain at the 2.3% operating point.
- Open Access
- 17 March 2026
- Published by Elsevier in SSRN Electronic Journal
Multi-Sensor Strain Measurement Data Processing and Fusion for Damage Identification in Aluminum Alloy Stiffened Plates: An Orthogonal Disentanglement Algorithm for Condition-Invariant Features under Varying Operating Conditions
In the field of precision strain measurement for structural integrity assessment, dynamic strain responses of aluminum alloy stiffened plates under varying operating conditions pose substantial challenges. Damage-related features are strongly coupled with measurement noise, operational variability, and inherent uncertainty, while state-of-the-art instrumentation frequently exhibits reduced accuracy and robustness due to insufficient data processing and fusion strategies that do not adequately control metrological uncertainty in noisy, multi-condition environments. To address these limitations, this paper proposes a novel data processing and fusion framework that designs an original orthogonal disentanglement mechanism as its core innovation. This mechanism is integrated with dual-scale perception for separating local damage-sensitive signals from global condition-driven variations, channel-attention-guided allocation to reduce interference, and adversarial learning to facilitate cross-domain alignment. This framework effectively mitigates feature entanglement while leveraging the spatial correlation among distributed sensors to enhance positioning accuracy and overall robustness. Validation on both simulated and experimental datasets under multiple noise levels demonstrates that the proposed method achieves superior accuracy, robustness, and metrological reliability compared with representative hybrid deep learning methods, indicating its strong potential for reliable damage identification in complex structural health monitoring scenarios.
- 17 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMulti-Sensor Strain Measurement Data Processing and Fusion for Damage Identification in Aluminum Alloy Stiffened Plates: An Orthogonal Disentanglement Algorithm for Condition-Invariant Features under Varying Operating Conditions
In the field of precision strain measurement for structural integrity assessment, dynamic strain responses of aluminum alloy stiffened plates under varying operating conditions pose substantial challenges. Damage-related features are strongly coupled with measurement noise, operational variability, and inherent uncertainty, while state-of-the-art instrumentation frequently exhibits reduced accuracy and robustness due to insufficient data processing and fusion strategies that do not adequately control metrological uncertainty in noisy, multi-condition environments. To address these limitations, this paper proposes a novel data processing and fusion framework that designs an original orthogonal disentanglement mechanism as its core innovation. This mechanism is integrated with dual-scale perception for separating local damage-sensitive signals from global condition-driven variations, channel-attention-guided allocation to reduce interference, and adversarial learning to facilitate cross-domain alignment. This framework effectively mitigates feature entanglement while leveraging the spatial correlation among distributed sensors to enhance positioning accuracy and overall robustness. Validation on both simulated and experimental datasets under multiple noise levels demonstrates that the proposed method achieves superior accuracy, robustness, and metrological reliability compared with representative hybrid deep learning methods, indicating its strong potential for reliable damage identification in complex structural health monitoring scenarios.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (16) , 1
A Wavenumber-Decoupled Damage Imaging Framework Based on a Guided Wave Sensor Network for Composite Structures
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (16) , 1
A Wavenumber-Decoupled Damage Imaging Framework Based on a Guided Wave Sensor Network for Composite Structures
- Open Access
High-Precision Pedestrian Indoor Positioning Method Based on Inertial and Magnetic Field Information
Long-term and high-precision positioning is the key to the pedestrian indoor positioning method. The estimation methods relying only on the inertial measurement unit (IMU) itself lack external observations that can provide absolute information, and the cumulative error easily leads to the distortion of the calculated trajectory. In this paper, based on the Extended Kalman Filter (EKF) algorithm, the environmental magnetic field information is taken as the external observation quantity, and a positioning method combining inertial navigation and the magnetic field is proposed. The cumulative error is suppressed from both the yaw angle and pedestrian pose, and the overall navigation and positioning accuracy is improved. The experimental results show that the proposed fusion method greatly improves the suppression of yaw angle and displacement errors. In a total distance of 297.08 m, the yaw angle error is reduced from 11.043° to 4.778°, and the position error is reduced from 8.999 m to 0.364 m. The relative average error decreases from 3.02% to 0.12%.
Open AccessHigh-Precision Pedestrian Indoor Positioning Method Based on Inertial and Magnetic Field Information
Long-term and high-precision positioning is the key to the pedestrian indoor positioning method. The estimation methods relying only on the inertial measurement unit (IMU) itself lack external observations that can provide absolute information, and the cumulative error easily leads to the distortion of the calculated trajectory. In this paper, based on the Extended Kalman Filter (EKF) algorithm, the environmental magnetic field information is taken as the external observation quantity, and a positioning method combining inertial navigation and the magnetic field is proposed. The cumulative error is suppressed from both the yaw angle and pedestrian pose, and the overall navigation and positioning accuracy is improved. The experimental results show that the proposed fusion method greatly improves the suppression of yaw angle and displacement errors. In a total distance of 297.08 m, the yaw angle error is reduced from 11.043° to 4.778°, and the position error is reduced from 8.999 m to 0.364 m. The relative average error decreases from 3.02% to 0.12%.
- 22 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-16
High-Precision Indoor Positioning Method Based on Multifeature Fusion of Inertial Sensor Network
Highly accurate positioning information can bring higher enhancement and value to basic applications. In the absence of infrastructure or third-party equipment, relying only on the inertial measurement unit (IMU) itself results in severe positioning drift. Aiming this challenge, we propose a high-precision localization method based on multi-feature fusion of body-area inertial sensor networks to achieve reliable and low-drift localization over long periods of time without the need for third-party assistance. The proposed method mainly includes three parts: the information of foot and shank are fused based on a ball hinge model to realize accurate and adaptive step length estimation under variable speed motion; the yaw angle is obtained by weighted fusion of inertial nodes mounted at the waist, shank, and foot; based on global multi-feature fusion with landmarks, the removal of stage historical errors of the position and location refinement are realized. The experimental results show that the proposed method can better suppress the dead reckoning error and effectively improve the positioning accuracy compared with the traditional single-IMU based dead reckoning method. Compared to another multi-IMU fusion approach, our method shows outperformance in the long-distance (1270 meters, 13 minutes). The average yaw angle error and average positioning error of the proposed method are 3.5 degrees and 3.3 meters, respectively, and it can better adapt to the variable speed movement.
- 22 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-16
High-Precision Indoor Positioning Method Based on Multifeature Fusion of Inertial Sensor Network
Highly accurate positioning information can bring higher enhancement and value to basic applications. In the absence of infrastructure or third-party equipment, relying only on the inertial measurement unit (IMU) itself results in severe positioning drift. Aiming this challenge, we propose a high-precision localization method based on multi-feature fusion of body-area inertial sensor networks to achieve reliable and low-drift localization over long periods of time without the need for third-party assistance. The proposed method mainly includes three parts: the information of foot and shank are fused based on a ball hinge model to realize accurate and adaptive step length estimation under variable speed motion; the yaw angle is obtained by weighted fusion of inertial nodes mounted at the waist, shank, and foot; based on global multi-feature fusion with landmarks, the removal of stage historical errors of the position and location refinement are realized. The experimental results show that the proposed method can better suppress the dead reckoning error and effectively improve the positioning accuracy compared with the traditional single-IMU based dead reckoning method. Compared to another multi-IMU fusion approach, our method shows outperformance in the long-distance (1270 meters, 13 minutes). The average yaw angle error and average positioning error of the proposed method are 3.5 degrees and 3.3 meters, respectively, and it can better adapt to the variable speed movement.
- 11 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (20) , 24696-24710
Mutual Calibration Method of Ring Laser Gyro and Optical Shaft Encoder
The calibration of a rotary table enables the quantitative evaluation of its performance and the traceability of angular motion parameters. However, there are few studies on the calibration method or equipment of rotary tables, especially for dynamic rotary tables. Existing measurement methods do not meet the requirements for rotary tables calibration in terms of measuring range, accuracy, feasibility, and simplicity of assembly. This paper presents a mutual calibration method based on a ring laser gyro and an optical shaft encoder. The scale factor of the ring laser gyro measurement model was calibrated based on the optical shaft encoder zero-position signal and the principle of full-circle closure. The ring laser gyro itself has the characteristics of high resolution and ideal circle segmentation. Therefore, the calibrated one is used to identify the parameters in the harmonic model established by the measurement error of the optical shaft encoder, and the measurement accuracy of the optical shaft encoder is improved by compensating for the error. Experimental results show that the mean measurement error of the ring laser gyro based on optical shaft encoder zero-position signal calibration is reduced by 72.38%, and the mean measurement error of the optical shaft encoder based on ring laser gyro measurement results compensation is reduced by 9.13%. The optical shaft encoder and the ring laser gyro are used to measure dynamic motion separately after mutual calibration, which makes the measuring system easy to install, debug, and operate, and has the advantages of wide range and high precision.
- 11 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (20) , 24696-24710
Mutual Calibration Method of Ring Laser Gyro and Optical Shaft Encoder
The calibration of a rotary table enables the quantitative evaluation of its performance and the traceability of angular motion parameters. However, there are few studies on the calibration method or equipment of rotary tables, especially for dynamic rotary tables. Existing measurement methods do not meet the requirements for rotary tables calibration in terms of measuring range, accuracy, feasibility, and simplicity of assembly. This paper presents a mutual calibration method based on a ring laser gyro and an optical shaft encoder. The scale factor of the ring laser gyro measurement model was calibrated based on the optical shaft encoder zero-position signal and the principle of full-circle closure. The ring laser gyro itself has the characteristics of high resolution and ideal circle segmentation. Therefore, the calibrated one is used to identify the parameters in the harmonic model established by the measurement error of the optical shaft encoder, and the measurement accuracy of the optical shaft encoder is improved by compensating for the error. Experimental results show that the mean measurement error of the ring laser gyro based on optical shaft encoder zero-position signal calibration is reduced by 72.38%, and the mean measurement error of the optical shaft encoder based on ring laser gyro measurement results compensation is reduced by 9.13%. The optical shaft encoder and the ring laser gyro are used to measure dynamic motion separately after mutual calibration, which makes the measuring system easy to install, debug, and operate, and has the advantages of wide range and high precision.
- 28 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (17) , 19117-19126
A New Generator of Pressure Pulsations With Wide Frequency and Amplitude
Time-varying pressure is measured in various fields and the calibration of pressure sensors for time-varying pressure is highly significant. The current piston-in-cylinder pressure pulsations generators generally operate at frequencies below 500 Hz. Due to the nonlinearity properties of gas compression and the limited displacement of the piston at high frequencies, generating sinusoidal pressure pulsations with large amplitudes becomes challenging when the frequency exceeds 500 Hz in a gaseous medium. This article proposes a new liquid-gas two-stage film generator with wide frequency range and amplitude range. First, the factors that affect the generation of high-frequency and high-amplitude pressure pulsations are analyzed. The primary factor is the limitation in piston displacement, resulting in small pressure amplitudes. A liquid-gas double-cylinder structure is proposed, which enhances the compression ratio by reducing the volume and outlet area of the secondary pneumatic cylinder. Second, the liquid-gas double-cylinder structure for pressure pulsation generators is further optimized. The symmetrical piston-in-cylinder generator produces pressure pulsations ranging from 45 to 500 Hz, while a liquid-gas two-stage film generator is utilized to generate pressure pulsations up to 2000 Hz. Experimental analyses were performed, indicating that the distortion of actual pressure pulsations is less than 12.5%, consistent with theoretical analysis. In the frequency range of 500–2000 Hz, the newly proposed pressure pulsation generator can reliably produce stable pressure pulsations at the specified frequencies.
- 28 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (17) , 19117-19126
A New Generator of Pressure Pulsations With Wide Frequency and Amplitude
Time-varying pressure is measured in various fields and the calibration of pressure sensors for time-varying pressure is highly significant. The current piston-in-cylinder pressure pulsations generators generally operate at frequencies below 500 Hz. Due to the nonlinearity properties of gas compression and the limited displacement of the piston at high frequencies, generating sinusoidal pressure pulsations with large amplitudes becomes challenging when the frequency exceeds 500 Hz in a gaseous medium. This article proposes a new liquid-gas two-stage film generator with wide frequency range and amplitude range. First, the factors that affect the generation of high-frequency and high-amplitude pressure pulsations are analyzed. The primary factor is the limitation in piston displacement, resulting in small pressure amplitudes. A liquid-gas double-cylinder structure is proposed, which enhances the compression ratio by reducing the volume and outlet area of the secondary pneumatic cylinder. Second, the liquid-gas double-cylinder structure for pressure pulsation generators is further optimized. The symmetrical piston-in-cylinder generator produces pressure pulsations ranging from 45 to 500 Hz, while a liquid-gas two-stage film generator is utilized to generate pressure pulsations up to 2000 Hz. Experimental analyses were performed, indicating that the distortion of actual pressure pulsations is less than 12.5%, consistent with theoretical analysis. In the frequency range of 500–2000 Hz, the newly proposed pressure pulsation generator can reliably produce stable pressure pulsations at the specified frequencies.
- Open Access
Methods for Comprehensive Calibration of a Low-Frequency Angular Acceleration Rotary Table
The total harmonic distortion (THD) index and its calculation methods are presented to calibrate the sinusoidal motion of the low-frequency angular acceleration rotary table (LFAART) and make up the incomprehensive evaluation based on the angular acceleration amplitude and frequency error indexes. The THD is calculated from two measurement schemes: a unique scheme combining the optical shaft encoder and the laser triangulation sensor and a regular scheme using the fiber optical gyroscope (FOG). An improved reversing moments recognition method is presented to upgrade the accuracy of solving the angular motion amplitude based on optical shaft encoder output. The field experiment shows that the difference in the THD values achieved using the combining scheme and FOG is within 0.11% when the signal-to-noise ratio of the FOG signal is higher than 7.7 dB, indicating the accuracy of the proposed methods and the feasibility of taking THD as the index.
Open AccessMethods for Comprehensive Calibration of a Low-Frequency Angular Acceleration Rotary Table
The total harmonic distortion (THD) index and its calculation methods are presented to calibrate the sinusoidal motion of the low-frequency angular acceleration rotary table (LFAART) and make up the incomprehensive evaluation based on the angular acceleration amplitude and frequency error indexes. The THD is calculated from two measurement schemes: a unique scheme combining the optical shaft encoder and the laser triangulation sensor and a regular scheme using the fiber optical gyroscope (FOG). An improved reversing moments recognition method is presented to upgrade the accuracy of solving the angular motion amplitude based on optical shaft encoder output. The field experiment shows that the difference in the THD values achieved using the combining scheme and FOG is within 0.11% when the signal-to-noise ratio of the FOG signal is higher than 7.7 dB, indicating the accuracy of the proposed methods and the feasibility of taking THD as the index.
- 26 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 70 (00189456) , 1-13
A Calibration Scheme With Combination of the Optical Shaft Encoder and Laser Triangulation Sensor for Low-Frequency Angular Acceleration Rotary Table
A low-frequency angular acceleration rotary table (LFAART) is capable of sinusoidal motion around a fixed position and essential to verify dynamic characteristics of a gyroscope device. The table’s performance at different swinging frequencies and amplitudes needs to be calibrated to realize traceability. However, there has been little research on the LFAART calibration. Besides, the existing measurement methods of sinusoidal angular motion parameters are not adequate for the needs of the LFAART calibration in terms of measurement range, precision, and feasibility. This article presents a calibration scheme combining an optical shaft encoder and a laser triangulation sensor. When the rotation amplitude covers sufficient slots of the scale grating, the optical shaft encoder is applied, while the laser triangulation sensor is used for a tiny amplitude motion. An algorithm is proposed to accurately calculate the angular acceleration amplitude and frequency of the rotation utilizing sinusoidal phase-modulated signals of the optical shaft encoder output. Singular zero-crossings of the series are discarded, and intervals of reversing moments are recognized according to the signal derivative’s sign at remaining zero-crossings. Then intervals are divided into subintervals to obtain accurate reversing moments. Finally, angular displacements between two adjacent reversing moments are estimated based on the fringe counting method, and the amplitude and frequency are calculated. The measurement uncertainty analysis and experiment demonstrate that the calibration scheme, including the algorithm, possesses high accuracy and resolution, wide range, and good operability, which perfectly solves the rotation’s angular acceleration amplitude and frequency.
- 26 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 70 (00189456) , 1-13
A Calibration Scheme With Combination of the Optical Shaft Encoder and Laser Triangulation Sensor for Low-Frequency Angular Acceleration Rotary Table
A low-frequency angular acceleration rotary table (LFAART) is capable of sinusoidal motion around a fixed position and essential to verify dynamic characteristics of a gyroscope device. The table’s performance at different swinging frequencies and amplitudes needs to be calibrated to realize traceability. However, there has been little research on the LFAART calibration. Besides, the existing measurement methods of sinusoidal angular motion parameters are not adequate for the needs of the LFAART calibration in terms of measurement range, precision, and feasibility. This article presents a calibration scheme combining an optical shaft encoder and a laser triangulation sensor. When the rotation amplitude covers sufficient slots of the scale grating, the optical shaft encoder is applied, while the laser triangulation sensor is used for a tiny amplitude motion. An algorithm is proposed to accurately calculate the angular acceleration amplitude and frequency of the rotation utilizing sinusoidal phase-modulated signals of the optical shaft encoder output. Singular zero-crossings of the series are discarded, and intervals of reversing moments are recognized according to the signal derivative’s sign at remaining zero-crossings. Then intervals are divided into subintervals to obtain accurate reversing moments. Finally, angular displacements between two adjacent reversing moments are estimated based on the fringe counting method, and the amplitude and frequency are calculated. The measurement uncertainty analysis and experiment demonstrate that the calibration scheme, including the algorithm, possesses high accuracy and resolution, wide range, and good operability, which perfectly solves the rotation’s angular acceleration amplitude and frequency.