Fangshuo Shi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Fangshuo Shi has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Wenbo Wang
State Key Laboratory of Photonics and Communications, School of Electronics, Peking University, Beijing, China
16 shared publications - Yanjun Chen
Hunan Key Laboratory of Nonferrous Resources and Geological Hazards Exploration, Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring, Ministry of Education, School of Geosciences and Info-Physics, Central South University, Changsha, China
16 shared publications - Zhengbin Li
State Key Laboratory of Photonics and Communications, School of Electronics, Peking University, Beijing, China
16 shared publications - Xinyu Cao
State Key Laboratory of Photonics and Communications, School of Electronics, Peking University, Beijing, China
15 shared publications - Lanxin Zhu
State Key Laboratory of Photonics and Communications, School of Electronics, Peking University, Beijing, China
14 shared publications - Huimin Huang
State Key Laboratory of Photonics and Communications, School of Electronics, Peking University, Beijing, China
12 shared publications
Latest Publications
- 4 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. PP (1530437X) , 1
Temperature-effects compensation of fiber-optic coils in fiber-optic gyroscope-based precision rotational sensing
Temperature effects induced into the fiber-optic coil degrade the precision of rotational sensing by fiber-optic gyroscopes. To address this issue, this study investigates how temperature-field characterization and the layout of temperature sensor measurement points influence temperature-error compensation under practical constraints. Under a limited number of temperature sensors, we find that, compared with measuring the temperature distribution along the fiber-optic coil, prioritizing the characterization of temperature differences between the inner and outer regions of the fiber-optic coil achieves more effective temperature-error compensation. Based on this insight, a multi-scale temperature field characterization and nonlinear modeling of temperature errors method is introduced to address temperature-induced error mechanisms. The method is validated through time-varying temperature variation experiments. Experimental results demonstrate that the optimized sensor layout improves the compensation performance by 6.8 dB. This work provides a systematic investigation of sensor-layout optimization and multi-scale feature construction for temperature compensation under practical constraints, enabling fiber-optic gyroscope-based precision rotational sensing with improved environmental adaptability.
- 4 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. PP (1530437X) , 1
Temperature-effects compensation of fiber-optic coils in fiber-optic gyroscope-based precision rotational sensing
Temperature effects induced into the fiber-optic coil degrade the precision of rotational sensing by fiber-optic gyroscopes. To address this issue, this study investigates how temperature-field characterization and the layout of temperature sensor measurement points influence temperature-error compensation under practical constraints. Under a limited number of temperature sensors, we find that, compared with measuring the temperature distribution along the fiber-optic coil, prioritizing the characterization of temperature differences between the inner and outer regions of the fiber-optic coil achieves more effective temperature-error compensation. Based on this insight, a multi-scale temperature field characterization and nonlinear modeling of temperature errors method is introduced to address temperature-induced error mechanisms. The method is validated through time-varying temperature variation experiments. Experimental results demonstrate that the optimized sensor layout improves the compensation performance by 6.8 dB. This work provides a systematic investigation of sensor-layout optimization and multi-scale feature construction for temperature compensation under practical constraints, enabling fiber-optic gyroscope-based precision rotational sensing with improved environmental adaptability.
- 9 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-9
Single-Point Transfer Function Method for Subsurface Structure Detection With a Six-Component Seismometer
We propose a single-station transfer function method using six-component (6-C) seismometer data. Conventional single-station methods assume monochromatic waves and rely on narrow bandpass filtering, which introduce spectral distortion and underestimate phase velocities. Through numerical simulations, we demonstrate that these issues arise from non-ideal filter responses and correlation contamination. To address these limitations, we reformulate the estimation as a system identification problem by modeling different seismic components as input–output pairs of a local system, and propose the single-point transfer function method. To enhance the accuracy of detection, we design a correlation-based slicing strategy that selects high-quality slices for analysis, and use a weighted average of slice-wise estimates as the final result. This framework is compatible with several existing single-station methods, offering deeper insight into wave–medium interactions. Experiments using ambient seismic noise demonstrate that the estimated shear-wave velocity profiles from both Love and Rayleigh waves show strong consistency with borehole drilling results, confirming the effectiveness of the proposed method.
- 9 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-9
Single-Point Transfer Function Method for Subsurface Structure Detection With a Six-Component Seismometer
We propose a single-station transfer function method using six-component (6-C) seismometer data. Conventional single-station methods assume monochromatic waves and rely on narrow bandpass filtering, which introduce spectral distortion and underestimate phase velocities. Through numerical simulations, we demonstrate that these issues arise from non-ideal filter responses and correlation contamination. To address these limitations, we reformulate the estimation as a system identification problem by modeling different seismic components as input–output pairs of a local system, and propose the single-point transfer function method. To enhance the accuracy of detection, we design a correlation-based slicing strategy that selects high-quality slices for analysis, and use a weighted average of slice-wise estimates as the final result. This framework is compatible with several existing single-station methods, offering deeper insight into wave–medium interactions. Experiments using ambient seismic noise demonstrate that the estimated shear-wave velocity profiles from both Love and Rayleigh waves show strong consistency with borehole drilling results, confirming the effectiveness of the proposed method.
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (22) , 41329-41336
An Interferometric Fiber-Optic Gyroscope With Low Self-Noise for Rotational Seismology Based on a Dual-Channel Configuration
In this work, we present the development of a high-performance interferometric fiber optic gyroscope specifically designed for rotational seismology. The system exhibits enhanced sensitivity and holds promise for precision sensing applications. A single broadband light source drives two Sagnac interferometers, each of which operates as a sensing channel. By employing reverse-phase modulation and a compensation scheme, the system effectively suppresses relative intensity noise. In addition, high-order eigenfrequency modulation is incorporated to mitigate the influence of thermal phase noise. The proposed configuration is implemented using two polarization-maintaining fiber coils, each 6.7 km in length and about 0.35 m in diameter. Experimental evaluations demonstrate strong noise correlation between channels and verify the significant improvements in sensitivity. The developed system achieves a self-noise level of 1.5 × 10 −9 rad/s/ √ Hz, which is 12 times lower than that of the single-channel system. The angular random walk and bias instability are measured at 3.8 × 10 −6 deg/ √ h and 3.7 × 10 −5 deg/h, respectively. This outstanding performance not only confirms the effectiveness of the proposed dual-channel strategy in mitigating intrinsic noise, but also validates the system’s suitability for high-precision geophysical applications.
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (22) , 41329-41336
An Interferometric Fiber-Optic Gyroscope With Low Self-Noise for Rotational Seismology Based on a Dual-Channel Configuration
In this work, we present the development of a high-performance interferometric fiber optic gyroscope specifically designed for rotational seismology. The system exhibits enhanced sensitivity and holds promise for precision sensing applications. A single broadband light source drives two Sagnac interferometers, each of which operates as a sensing channel. By employing reverse-phase modulation and a compensation scheme, the system effectively suppresses relative intensity noise. In addition, high-order eigenfrequency modulation is incorporated to mitigate the influence of thermal phase noise. The proposed configuration is implemented using two polarization-maintaining fiber coils, each 6.7 km in length and about 0.35 m in diameter. Experimental evaluations demonstrate strong noise correlation between channels and verify the significant improvements in sensitivity. The developed system achieves a self-noise level of 1.5 × 10 −9 rad/s/ √ Hz, which is 12 times lower than that of the single-channel system. The angular random walk and bias instability are measured at 3.8 × 10 −6 deg/ √ h and 3.7 × 10 −5 deg/h, respectively. This outstanding performance not only confirms the effectiveness of the proposed dual-channel strategy in mitigating intrinsic noise, but also validates the system’s suitability for high-precision geophysical applications.
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (22) , 41337-41344
Reciprocal Fiber-Optic Differential Interferometer Based on Hybrid Modulation
The fiber-optic differential interferometer (FODI) has emerged as a high-sensitivity sensing technique capable of directly measuring the temporal derivatives of physical signals. In this study, we propose and experimentally validate a reciprocal FODI system that employs a hybrid modulation scheme. A 1-km fiber delay line is used in static and dynamic experiments to evaluate the performance of the proposed FODI. In static tests, the system achieves a phase sensitivity of 2.6 × 10–6 rad/ √ Hz within a wideband from 1 mHz to 100 Hz. In dynamic tests, the system demonstrates a dynamic range exceeding 103 dB and achieves a total harmonic distortion as low as –69 dB under maximum input conditions. Furthermore, amplitude response tests yield an R-square value exceeding 0.9999. These results confirm that the proposed FODI system offers balanced comprehensive performance in sensitivity, stability, dynamic response, and linearity, providing an enhanced tool for applications such as strain rate sensing and vibration analysis.
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (22) , 41337-41344
Reciprocal Fiber-Optic Differential Interferometer Based on Hybrid Modulation
The fiber-optic differential interferometer (FODI) has emerged as a high-sensitivity sensing technique capable of directly measuring the temporal derivatives of physical signals. In this study, we propose and experimentally validate a reciprocal FODI system that employs a hybrid modulation scheme. A 1-km fiber delay line is used in static and dynamic experiments to evaluate the performance of the proposed FODI. In static tests, the system achieves a phase sensitivity of 2.6 × 10–6 rad/ √ Hz within a wideband from 1 mHz to 100 Hz. In dynamic tests, the system demonstrates a dynamic range exceeding 103 dB and achieves a total harmonic distortion as low as –69 dB under maximum input conditions. Furthermore, amplitude response tests yield an R-square value exceeding 0.9999. These results confirm that the proposed FODI system offers balanced comprehensive performance in sensitivity, stability, dynamic response, and linearity, providing an enhanced tool for applications such as strain rate sensing and vibration analysis.
- 9 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 43 (22) , 10295-10301
Low-Cost High-Efficiency Brillouin Random Fiber Laser Based on Coherent Self-Injection Locking
We realize a narrow-linewidth Brillouin random fiber laser (BRFL) that uses the proposed coherent self-injection locking with a semi-open linear cavity configuration. The proposed laser demonstrates balanced performance, including a low threshold power of 9.22 mW, high laser efficiency of 20%, and exceptional optical signal-to-noise ratio (OSNR) of 67.01 dB. Moreover, the linewidth is compressed from the hundred-kilohertz level to the hundred-hertz level, achieving a 3 dB linewidth of 412 Hz, which corresponds to a 1214 times compression relative to the pump laser. Statistical analyses confirmed that multimode resonances are effectively suppressed, thereby significantly enhancing single-mode stability. Furthermore, frequency noise measurements confirmed significant noise suppression across the entire tested frequency spectrum. These results highlight the potential of coherent self-injection locking for developing highly stable, efficient, and narrow-linewidth BRFLs, making them promising for precision optical applications.
- 9 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 43 (22) , 10295-10301
Low-Cost High-Efficiency Brillouin Random Fiber Laser Based on Coherent Self-Injection Locking
We realize a narrow-linewidth Brillouin random fiber laser (BRFL) that uses the proposed coherent self-injection locking with a semi-open linear cavity configuration. The proposed laser demonstrates balanced performance, including a low threshold power of 9.22 mW, high laser efficiency of 20%, and exceptional optical signal-to-noise ratio (OSNR) of 67.01 dB. Moreover, the linewidth is compressed from the hundred-kilohertz level to the hundred-hertz level, achieving a 3 dB linewidth of 412 Hz, which corresponds to a 1214 times compression relative to the pump laser. Statistical analyses confirmed that multimode resonances are effectively suppressed, thereby significantly enhancing single-mode stability. Furthermore, frequency noise measurements confirmed significant noise suppression across the entire tested frequency spectrum. These results highlight the potential of coherent self-injection locking for developing highly stable, efficient, and narrow-linewidth BRFLs, making them promising for precision optical applications.
- 11 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (18) , 34655-34662
Simultaneous Self-Contained Temperature and Angular Velocity Measurement in Interferometric Fiber-Optic Gyroscopes
Interferometric fiber-optic gyroscopes (IFOGs) are sensitive to ambient temperature variations, which can introduce drift in the measured angular velocity. To mitigate such effects, accurate access to real-time temperature information is essential. This article proposes a self-contained multiparameter measurement method that enables simultaneous acquisition of temperature and angular velocity without relying on external sensors. The approach utilizes a multiplexing four-state modulation scheme to embed temperature-related information in the output signal, which is then extracted and used to adjust the signal processing accordingly. Both simulation and experimental results confirm the effectiveness of the proposed method across a wide thermal range, with temperature measurement achieving a root-mean-square error (RMSE) of$0.3722~^{\circ } $ C. Temperature correction is carried out using the real-time measured temperature, and experimental results demonstrate improved stability, with the 100-s averaged standard deviation reduced from$0.252~^{\circ } $ /h to$0.023~^{\circ } $ /h by a factor of 10.9, and the maximum deviation suppressed from$0.819~^{\circ } $ /h to$0.097~^{\circ } $ /h by a factor of 8.4. As the method operates entirely within the existing IFOG modulation and detection framework, it offers a practical solution for improving system performance in thermally dynamic environments.- 11 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (18) , 34655-34662
Simultaneous Self-Contained Temperature and Angular Velocity Measurement in Interferometric Fiber-Optic Gyroscopes
Interferometric fiber-optic gyroscopes (IFOGs) are sensitive to ambient temperature variations, which can introduce drift in the measured angular velocity. To mitigate such effects, accurate access to real-time temperature information is essential. This article proposes a self-contained multiparameter measurement method that enables simultaneous acquisition of temperature and angular velocity without relying on external sensors. The approach utilizes a multiplexing four-state modulation scheme to embed temperature-related information in the output signal, which is then extracted and used to adjust the signal processing accordingly. Both simulation and experimental results confirm the effectiveness of the proposed method across a wide thermal range, with temperature measurement achieving a root-mean-square error (RMSE) of$0.3722~^{\circ } $ C. Temperature correction is carried out using the real-time measured temperature, and experimental results demonstrate improved stability, with the 100-s averaged standard deviation reduced from$0.252~^{\circ } $ /h to$0.023~^{\circ } $ /h by a factor of 10.9, and the maximum deviation suppressed from$0.819~^{\circ } $ /h to$0.097~^{\circ } $ /h by a factor of 8.4. As the method operates entirely within the existing IFOG modulation and detection framework, it offers a practical solution for improving system performance in thermally dynamic environments.- 28 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 43 (18) , 8966-8973
Dual-Channel Fiber Optic Differential Interferometer With Square-Wave Modulation for Advanced Sensing Applications
The fiber optic differential interferometer (FODI) is a sensor designed to directly detect the temporal derivatives of physical quantities. We propose a novel dual-channel FODI (DC-FODI) with square-wave modulation, which enhances sensitivity and improves the dynamic range compared with previous designs. This system uses two optical fibers for simultaneous detection, each embedded within a Sagnac interferometer. The fiber delay lines in both interferometers are of equal length. A pair of equal-amplitude, reverse-phase square wave signals modulates the two interferometers. In this study, the mechanism for suppressing relative intensity noise (RIN) is thoroughly analyzed. In addition, the two delay lines are wound into coils of equal diameter and oriented in opposite directions to mitigate parasitic phase shifts caused by the Earth's rotation and environmental vibrations. The DC-FODI employs two 1-km optical fibers as delay lines. In static tests, DC-FODI demonstrates an over 20-fold improvement in phase sensitivity compared with the single-channel system, with the phase sensitivity estimated at $3 \times 10^{-8} \, \text{rad}/\sqrt{\text{Hz}}$. The system exhibits an essentially flat self-noise spectrum within the frequency range from 0.002 Hz to 100 Hz. Dynamic experiments confirm its sensing capability, demonstrating a dynamic range exceeding 112 dB. High-performance FODI is expected to drive advancements in measurement fields such as acceleration and strain rate measurement.
- 28 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 43 (18) , 8966-8973
Dual-Channel Fiber Optic Differential Interferometer With Square-Wave Modulation for Advanced Sensing Applications
The fiber optic differential interferometer (FODI) is a sensor designed to directly detect the temporal derivatives of physical quantities. We propose a novel dual-channel FODI (DC-FODI) with square-wave modulation, which enhances sensitivity and improves the dynamic range compared with previous designs. This system uses two optical fibers for simultaneous detection, each embedded within a Sagnac interferometer. The fiber delay lines in both interferometers are of equal length. A pair of equal-amplitude, reverse-phase square wave signals modulates the two interferometers. In this study, the mechanism for suppressing relative intensity noise (RIN) is thoroughly analyzed. In addition, the two delay lines are wound into coils of equal diameter and oriented in opposite directions to mitigate parasitic phase shifts caused by the Earth's rotation and environmental vibrations. The DC-FODI employs two 1-km optical fibers as delay lines. In static tests, DC-FODI demonstrates an over 20-fold improvement in phase sensitivity compared with the single-channel system, with the phase sensitivity estimated at $3 \times 10^{-8} \, \text{rad}/\sqrt{\text{Hz}}$. The system exhibits an essentially flat self-noise spectrum within the frequency range from 0.002 Hz to 100 Hz. Dynamic experiments confirm its sensing capability, demonstrating a dynamic range exceeding 112 dB. High-performance FODI is expected to drive advancements in measurement fields such as acceleration and strain rate measurement.
- 21 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-10
Joint Temperature Drift Compensation for Dual-Channel Fiber-Optic Gyroscopes
This paper proposes a joint temperature drift compensation method for dual-channel interferometric fiber-optic gyroscopes (IFOGs) to mitigate the impact of temperature variations on modulation depth and measurement performance. The proposed method establishes a temperature-dependent modulation depth model and develops a real-time compensation strategy that dynamically adjusts the modulation voltage to maintain optimal modulation depth. To ensure consistency between the two sensing channels, a joint compensation method is implemented, incorporating power balancing and residual correction. Experimental results demonstrate that this method effectively suppresses temperature-induced drift, improving sensitivity by 2.5 times, expanding the detection bandwidth from 0.3 Hz–50 Hz to 0.01 Hz–50 Hz, and enhancing bias instability (BI) by over 2.7 times. By significantly improving the detection bandwidth and stability of dual-channel IFOGs under temperature variations, this compensation method offers a promising solution for high-precision measurement applications.
- 21 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-10
Joint Temperature Drift Compensation for Dual-Channel Fiber-Optic Gyroscopes
This paper proposes a joint temperature drift compensation method for dual-channel interferometric fiber-optic gyroscopes (IFOGs) to mitigate the impact of temperature variations on modulation depth and measurement performance. The proposed method establishes a temperature-dependent modulation depth model and develops a real-time compensation strategy that dynamically adjusts the modulation voltage to maintain optimal modulation depth. To ensure consistency between the two sensing channels, a joint compensation method is implemented, incorporating power balancing and residual correction. Experimental results demonstrate that this method effectively suppresses temperature-induced drift, improving sensitivity by 2.5 times, expanding the detection bandwidth from 0.3 Hz–50 Hz to 0.01 Hz–50 Hz, and enhancing bias instability (BI) by over 2.7 times. By significantly improving the detection bandwidth and stability of dual-channel IFOGs under temperature variations, this compensation method offers a promising solution for high-precision measurement applications.
- 12 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-17
Rotation Polarization Analysis With Single-Station Six-Component Seismometer
Accurate wave component separation is essential for seismic signal processing, playing a key role in seismic source localization and dispersion curve inversion. However, traditional three-component polarization filters struggle to distinguish various wave components due to complex wave propagation and ambient noise. Current single-station six-component methods emphasize wave type identification but lack detailed time-frequency information, limiting their applicability in certain scenarios. To address these limitations, we present a novel algorithm that effectively isolates circularly polarized components using single-station six-component data, offering detailed polarization parameters across time and frequency domains. To validate our algorithm, we conducted a series of experiments, using back azimuth estimation as a key performance metric. Compared to the traditional 3-C method, the proposed algorithm reduces the mean absolute error by 79.81% and the standard deviation by 83.07%. These improvements demonstrate its superior accuracy in distinguishing various wave components and its enhanced convergence in back azimuth estimation. Additionally, it achieves comparable performance with shorter data while providing more time-frequency domain information. These results confirm the advanced capability of the algorithm to accurately isolate different wave components.
- 12 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-17
Rotation Polarization Analysis With Single-Station Six-Component Seismometer
Accurate wave component separation is essential for seismic signal processing, playing a key role in seismic source localization and dispersion curve inversion. However, traditional three-component polarization filters struggle to distinguish various wave components due to complex wave propagation and ambient noise. Current single-station six-component methods emphasize wave type identification but lack detailed time-frequency information, limiting their applicability in certain scenarios. To address these limitations, we present a novel algorithm that effectively isolates circularly polarized components using single-station six-component data, offering detailed polarization parameters across time and frequency domains. To validate our algorithm, we conducted a series of experiments, using back azimuth estimation as a key performance metric. Compared to the traditional 3-C method, the proposed algorithm reduces the mean absolute error by 79.81% and the standard deviation by 83.07%. These improvements demonstrate its superior accuracy in distinguishing various wave components and its enhanced convergence in back azimuth estimation. Additionally, it achieves comparable performance with shorter data while providing more time-frequency domain information. These results confirm the advanced capability of the algorithm to accurately isolate different wave components.
- 2 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (13) , 25861-25871
ResNet-Based Single-Station Localization Method Based on Six-Component Seismometer
The adoption of intelligent mining systems improves operation safety and efficiency, requiring accurate localization of shearers in underground environments. However, traditional GPS-based localization methods are ineffective due to underground conditions. Alternative technologies often require receiver arrays or additional signal generators, which can suffer from unreliable connections and extensive hardware deployment. To address these challenges, we propose a novel single-station method that utilizes seismic signals generated by the shearer. This method simplifies the requirements for signal source by analyzing continuous oscillations produced during shearer operation, and simplifies hardware implementation using a single six-component (6-C) seismometer to record these oscillations as time series data. Given the presence of ambient noise, we employ a Residual Network (ResNet) to extract relevant features from the seismic data. As a deep learning architecture, ResNet incorporates residual blocks to mitigate the vanishing gradient problem, enhancing feature extraction performance. Experimental results demonstrate the effectiveness of the proposed method. Using oscillation data collected during shearer operation to train the ResNet model, we achieve a prediction accuracy of 94.89%, outperforming traditional single-station methods such as cross-correlation (CC) and 6-C multiple signal classification (MUSIC). This study demonstrates that our method does not require signal generators or extensive sensor deployments. By leveraging seismic data and deep learning techniques, significant enhancements in shearer localization can be achieved, contributing to the development of safer and more efficient mining operations.
- 2 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (13) , 25861-25871
ResNet-Based Single-Station Localization Method Based on Six-Component Seismometer
The adoption of intelligent mining systems improves operation safety and efficiency, requiring accurate localization of shearers in underground environments. However, traditional GPS-based localization methods are ineffective due to underground conditions. Alternative technologies often require receiver arrays or additional signal generators, which can suffer from unreliable connections and extensive hardware deployment. To address these challenges, we propose a novel single-station method that utilizes seismic signals generated by the shearer. This method simplifies the requirements for signal source by analyzing continuous oscillations produced during shearer operation, and simplifies hardware implementation using a single six-component (6-C) seismometer to record these oscillations as time series data. Given the presence of ambient noise, we employ a Residual Network (ResNet) to extract relevant features from the seismic data. As a deep learning architecture, ResNet incorporates residual blocks to mitigate the vanishing gradient problem, enhancing feature extraction performance. Experimental results demonstrate the effectiveness of the proposed method. Using oscillation data collected during shearer operation to train the ResNet model, we achieve a prediction accuracy of 94.89%, outperforming traditional single-station methods such as cross-correlation (CC) and 6-C multiple signal classification (MUSIC). This study demonstrates that our method does not require signal generators or extensive sensor deployments. By leveraging seismic data and deep learning techniques, significant enhancements in shearer localization can be achieved, contributing to the development of safer and more efficient mining operations.