Xingye Bai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xingye Bai has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
Jun Lin
3 shared publications- Jun Lin
Jilin University
3 shared publications - Xin Zhao
State Key Laboratory of Deep Earth Exploration and Imaging and the College of Instrumentation Science and Electrical Engineering, Jilin University, Changchun, China
3 shared publications - Chuandong Jiang
Jilin University
2 shared publications - Fudong Zhang
College of Instrumentation and Electrical Engineering, Jilin University, Changchun, China
2 shared publications Tianxiong Li
Key Laboratory of Geophysical Exploration Equipment (Jilin University), Ministry of Education of China
2 shared publications
Latest Publications
- 17 December 2025
- Published by Optica Publishing Group in Optics Express
- Vol. 33 (26) , 54035-54051
Unified self-training framework for open-set semi-supervised event recognition in distributed optical fiber sensing
Distributed optical fiber sensing (DOFS) has become a prominent solution for real-time and wide-area monitoring owing to its long sensing range and high sensitivity. However, existing event recognition approaches rely heavily on large-scale annotated datasets, making them difficult to deploy in practical applications where labeled data are scarce and unknown events frequently occur. Moreover, most current models are designed for closed environments and struggle to identify unseen disturbances under real-world conditions. To address these challenges, we propose an open-set semi-supervised event recognition method for DOFS systems based on a unified self-training framework. The proposed method selects reliable unknown samples using a buffer-based selection strategy and integrates closed-set and open-set self-training into a unified optimization process, enabling simultaneous recognition of known and unseen events. Experimental results show that our method delivers strong and consistent performance in both close-set and open-set event recognition scenarios, achieving 95.14% open-set accuracy and substantially outperforming representative baselines such as IOMatch and OpenMatch under low-supervision conditions (1% labeled data). This demonstrates that the proposed method markedly improves the recognition performance and robustness of DOFS systems, providing a reliable foundation for intelligent optical fiber sensing in open and dynamic environments.
- 17 December 2025
- Published by Optica Publishing Group in Optics Express
- Vol. 33 (26) , 54035-54051
Unified self-training framework for open-set semi-supervised event recognition in distributed optical fiber sensing
Distributed optical fiber sensing (DOFS) has become a prominent solution for real-time and wide-area monitoring owing to its long sensing range and high sensitivity. However, existing event recognition approaches rely heavily on large-scale annotated datasets, making them difficult to deploy in practical applications where labeled data are scarce and unknown events frequently occur. Moreover, most current models are designed for closed environments and struggle to identify unseen disturbances under real-world conditions. To address these challenges, we propose an open-set semi-supervised event recognition method for DOFS systems based on a unified self-training framework. The proposed method selects reliable unknown samples using a buffer-based selection strategy and integrates closed-set and open-set self-training into a unified optimization process, enabling simultaneous recognition of known and unseen events. Experimental results show that our method delivers strong and consistent performance in both close-set and open-set event recognition scenarios, achieving 95.14% open-set accuracy and substantially outperforming representative baselines such as IOMatch and OpenMatch under low-supervision conditions (1% labeled data). This demonstrates that the proposed method markedly improves the recognition performance and robustness of DOFS systems, providing a reliable foundation for intelligent optical fiber sensing in open and dynamic environments.
- 1 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-12
SNR Enhancement Method for Φ-OTDR System Based on Sdr SampEn-BiPSO-SVD Algorithm
To solve the problem of a low signal-to-noise ratio (SNR) of the vibration positioning curve in the phase-sensitive optical time-domain reflectometer ($\Phi $ -OTDR), a denoising algorithm based on spatial dependence recurrence sample entropy (Sdr SampEn) and singular value decomposition (SVD) optimized by improved particle swarm optimization algorithm based on the Beer Antennae search (BiPSO) algorithm (Sdr SampEn-BiPSO-SVD) is proposed. First, the$\Phi $ -OTDR system based on a heterodyne coherent detection structure is built, the vibration signal is obtained by the I/Q quadrature demodulation algorithm, and the noise source of the vibration signal is analyzed. Second, the mathematical principle and the denoising process of the Sdr SampEn-BiPSO-SVD algorithm is established. Finally, the noise suppression capability of Sdr SampEn-BiPSO-SVD algorithm is tested by simulation experiments and$\Phi $ -OTDR system experiments, and compared with wavelet, empirical mode decomposition combined with Pearson correlation coefficient (EMD-PCC), variational mode decomposition (VMD). The experimental results show that the SNR of vibration positioning curves at 1 Hz, 100 Hz, 500 Hz, and 1 kHz improved to 21.28, 29.08, 16.55, and 32.99 dB, respectively, and demonstrate the ability to identify multipoint vibration and improve the quality of multipoint vibration positioning signals. The feasibility and effectiveness of the algorithm have been proven in practical applications. This algorithm provides a new means for the application of the$\Phi $ -OTDR system in the field of seismic exploration.- 1 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-12
SNR Enhancement Method for Φ-OTDR System Based on Sdr SampEn-BiPSO-SVD Algorithm
To solve the problem of a low signal-to-noise ratio (SNR) of the vibration positioning curve in the phase-sensitive optical time-domain reflectometer ($\Phi $ -OTDR), a denoising algorithm based on spatial dependence recurrence sample entropy (Sdr SampEn) and singular value decomposition (SVD) optimized by improved particle swarm optimization algorithm based on the Beer Antennae search (BiPSO) algorithm (Sdr SampEn-BiPSO-SVD) is proposed. First, the$\Phi $ -OTDR system based on a heterodyne coherent detection structure is built, the vibration signal is obtained by the I/Q quadrature demodulation algorithm, and the noise source of the vibration signal is analyzed. Second, the mathematical principle and the denoising process of the Sdr SampEn-BiPSO-SVD algorithm is established. Finally, the noise suppression capability of Sdr SampEn-BiPSO-SVD algorithm is tested by simulation experiments and$\Phi $ -OTDR system experiments, and compared with wavelet, empirical mode decomposition combined with Pearson correlation coefficient (EMD-PCC), variational mode decomposition (VMD). The experimental results show that the SNR of vibration positioning curves at 1 Hz, 100 Hz, 500 Hz, and 1 kHz improved to 21.28, 29.08, 16.55, and 32.99 dB, respectively, and demonstrate the ability to identify multipoint vibration and improve the quality of multipoint vibration positioning signals. The feasibility and effectiveness of the algorithm have been proven in practical applications. This algorithm provides a new means for the application of the$\Phi $ -OTDR system in the field of seismic exploration.- 8 September 2023
- Published by Optica Publishing Group in Optics Express
- Vol. 31 (19) , 31463-31485
Phase drift and noise suppression method based on SEE-SGMD-PCC in a distributed acoustic sensor
The problems of laser source frequency drift (LSFD) and phase noise in distributed acoustic sensor (DAS) make it difficult to recover the true vibration phase curve, which limits its application scope in the field of seismic exploration. In order to suppress the influence of phase noise and phase drift on the phase signal and improve the signal-to-noise ratio (SNR) of the phase curve, this paper proposes a method based on symmetric extreme value expansion, symplectic geometry mode decomposition and Pearson correlation coefficient (SEE-SGMD-PCC). Firstly, the mathematical principles and processing flow of the SEE-SGMD-PCC algorithm were introduced, and the effectiveness of this method was verified through multiple sets of simulation experiments. Secondly, The DAS system implemented using digital heterodyne coherent detection technology was used as the experimental platform, and the phase drift and phase noise sources of the phase signal were analyzed in detail. Next, in single frequency and multi frequency vibration signal experiments, compared with other methods, the SNR of the phase signal is significantly improved, and the phase information is effectively restored. Finally, the feasibility of the proposed method was demonstrated through two on-site experiments. The proposal of this method further promotes the application process of DAS in complex seismic exploration environments.(c) 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
- 8 September 2023
- Published by Optica Publishing Group in Optics Express
- Vol. 31 (19) , 31463-31485
Phase drift and noise suppression method based on SEE-SGMD-PCC in a distributed acoustic sensor
The problems of laser source frequency drift (LSFD) and phase noise in distributed acoustic sensor (DAS) make it difficult to recover the true vibration phase curve, which limits its application scope in the field of seismic exploration. In order to suppress the influence of phase noise and phase drift on the phase signal and improve the signal-to-noise ratio (SNR) of the phase curve, this paper proposes a method based on symmetric extreme value expansion, symplectic geometry mode decomposition and Pearson correlation coefficient (SEE-SGMD-PCC). Firstly, the mathematical principles and processing flow of the SEE-SGMD-PCC algorithm were introduced, and the effectiveness of this method was verified through multiple sets of simulation experiments. Secondly, The DAS system implemented using digital heterodyne coherent detection technology was used as the experimental platform, and the phase drift and phase noise sources of the phase signal were analyzed in detail. Next, in single frequency and multi frequency vibration signal experiments, compared with other methods, the SNR of the phase signal is significantly improved, and the phase information is effectively restored. Finally, the feasibility of the proposed method was demonstrated through two on-site experiments. The proposal of this method further promotes the application process of DAS in complex seismic exploration environments.(c) 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
- 22 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (19) , 22608-22619
Fading Noise Suppression Method of Φ-OTDR System Based on GA-VMD Algorithm
A distributed optical fiber acoustic sensing system (DAS) based on phase-sensitive optical time domain reflectometry (Φ-OTDR) can detect the vibration information along an optical fiber, and has been widely used for various engineering detections. However, the high-coherence detection light source used in a DAS will cause coherent fading in the fiber, leading to system fading noise and affecting the monitoring of the target vibration events. In order to solve the problem of the fading noise in the DAS, this study considered a method of suppressing the fading noise based on genetic algorithm (GA) and variational mode decomposition (VMD). Furthermore, the parameter selection problem in the VMD method is improved by introducing the GA. By introducing sample entropy as fitness function and combining it with the GA, this method adaptively searches for the optimal decomposition parameters by iterating the minimum value of the fitness function according to the type of input signal. Then, it decomposes the noisy DAS data into multiple intrinsic mode functions (IMFs), and the permutation entropy and time-frequency domain information of each component were used to extract the effective IMF components for signal reconstruction to obtain the denoised signal. Experimental results show that compared with EMD, EEMD and CEEMD methods, the proposed method can eliminate the mode aliasing problem, effectively suppress the fading noise of the DAS, improve the signal-to-noise ratio, and reduce the root-mean-square error. Thus, it is possibly to more accurately monitor vibration events, improving the practical application value of the DAS.
- 22 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (19) , 22608-22619
Fading Noise Suppression Method of Φ-OTDR System Based on GA-VMD Algorithm
A distributed optical fiber acoustic sensing system (DAS) based on phase-sensitive optical time domain reflectometry (Φ-OTDR) can detect the vibration information along an optical fiber, and has been widely used for various engineering detections. However, the high-coherence detection light source used in a DAS will cause coherent fading in the fiber, leading to system fading noise and affecting the monitoring of the target vibration events. In order to solve the problem of the fading noise in the DAS, this study considered a method of suppressing the fading noise based on genetic algorithm (GA) and variational mode decomposition (VMD). Furthermore, the parameter selection problem in the VMD method is improved by introducing the GA. By introducing sample entropy as fitness function and combining it with the GA, this method adaptively searches for the optimal decomposition parameters by iterating the minimum value of the fitness function according to the type of input signal. Then, it decomposes the noisy DAS data into multiple intrinsic mode functions (IMFs), and the permutation entropy and time-frequency domain information of each component were used to extract the effective IMF components for signal reconstruction to obtain the denoised signal. Experimental results show that compared with EMD, EEMD and CEEMD methods, the proposed method can eliminate the mode aliasing problem, effectively suppress the fading noise of the DAS, improve the signal-to-noise ratio, and reduce the root-mean-square error. Thus, it is possibly to more accurately monitor vibration events, improving the practical application value of the DAS.