YDYunke Du

Yunke Du

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yunke Du has a total of 19 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • An Enhanced Multiple Intrusion Event Localization Algorithm Based on Dual MZ–Sagnac Optical Fiber Sensing Perimeter Security System

    by Yang Yan, Churui Li, Rui Jin, Chenyue He, Yuzhe Sun, Yunke Du, Chao Wang

    Interferometric distributed optical fiber vibration sensing technology has been widely applied in the field of perimeter security. Among various structures, the dual MZ–Sagnac optical fiber sensing perimeter security system (the dual MZ–Sagnac system) plays an important role in perimeter security due to its high sensitivity, electromagnetic resistance, corrosion resistance, high stability and low cost. However, under the condition of low-SNR signals, the traditional event localization method for the dual MZ–Sagnac system, which directly averages the localization results of continuous short-interval signals, leads to great errors. In this paper, by employing the Cram´er-Rao Lower Bound (CRLB) theory of cross-correlation time delay estimation method, a Weighted Gaussian Superposition (WGS) algorithm based on the dual MZ–Sagnac system is proposed to enhance event localization accuracy and distinguish multiple vibration events. Vibration signals are collected and CRLB conformity to the dual MZ–Sagnac system is validated. Furthermore, the WGS algorithm reveals a minimum MSE of 1.63, corresponding to the error of 1.28 m and an accuracy improvement of 72 % in the experiment. Additionally, an application is performed to evaluate the effectiveness of the WGS algorithm for multiple events discrimination and localization. In perimeter security scenarios utilizing the dual MZ-Sagnac system, the WGS algorithm not only further refines the localization precision but also enables the system with multiple events discrimination, allowing the measurement instrumentation to better adapt to engineering applications.

  • Interference Module Noise Suppression Model for Interferometric Distributed Optical Fiber Sensing

    by Yunke Du, Yang Yan, Churui Li, Bo Ma, Chao Wang, Bo Jia

    Interferometric distributed optical fiber vibration sensing (iDOFVS) system is critically important in optical fiber sensing. The stability of interference module environment significantly impacts system noise performance. In this paper, an interference-module time-delay fiber (TDF) noise suppression model is proposed for iDOFVS on the basis of an auxiliary reference interferometer, which addresses persistent localization failure under TDF-induced noise. Through theoretical analysis, the transmission principle of the interference-module TDF-induced noise is clarified and validated. Vibration signal localization experiments are conducted over a 50 km urban optical fiber line. Experimental results show that the proposed technique reduces the overall localization error from 61.3 m to 18.9 m and increases the average signal-to-noise ratio from 20.25 dB to 26.81 dB. The proposed technique improves localization accuracy while reducing the size of the equipment, offering practical value for distributed vibration sensing applications.

  • High-Accuracy CaAlg-Coated D-Shaped Optical Fiber Humidity Sensor Using Multipeak Wavelength Shift Averaging Method

    by Yuzhe Sun, Yunke Du, Zongrui Ma, Ruixin Liang, Rui Jin, Yang Yan, Yongfeng Mei, Dong Zhao, Chao Wang, Bo Jia

    A calcium alginate (CaAlg)-coated D-shaped single mode fiber humidity sensor is proposed in this paper. The sensor is fabricated by grinding single mode fiber and wrapping CaAlg with rolling stress. The D-shaped optical fiber sensor utilizes outward cladding mode to enhance sensing sensitivity to external refractive index changes. To visualize the spectrum response of D-shaped optical fiber humidity sensor, the periodic resonance spectrum is generated by tapered fiber connected to a broadband light source. External humidity changes are detected by observing transmission spectrum wavelength shift. The multi-peak wavelength shift averaging method is used to improve the humidity measurement linear accuracy. Results show that the relative humidity (RH) sensitivity is 0.0269 nm/%RH in 10-99%RH range, with temperature insensitive. This study provides an economical method for achieving accurate humidity detection using relatively low resolution commercial optical spectrometers. With high humidity sensing sensitivity, measurement precision, and temperature insensitivity, the proposed sensor has potential applications in practical humidity measurement.


  • A Novel Unidirectional Asymmetric Distributed Interferometer With Forward-Propagating Light for Ultra-Long Submarine Optical Cable Security Monitoring

    by Yang Yan, Chenyue He, Churui Li, Yunke Du, Rui Jin, Yuzhe Sun, Ruixin Liang, Zongrui Ma, Chao Wang, Bo Jia

    Submarine optical cables play an important role in modern global communication system. However, existing monitoring technologies are unable to guarantee the security of ultra-long submarine optical cables. In this paper, a novel dual-wavelength unidirectional circular propagation asymmetric Mach-Zehnder interferometer (UCAMZI) is proposed. By utilizing existing commercial submarine repeaters, UCAMZI can achieve ultra-long-distance sensing in submarine optical fiber security monitoring, and shows excellent dynamic range and wide frequency response. The UCAMZI is verified in a laboratory environment with a sensing distance of 150 km, achieving a root mean square error (RMSE) of approximately 171 m along the entire sensing fiber, which demonstrates localization accuracy. The frequency response range of UCAMZI is also tested, exhibiting stable performance within 5 kHz to 37 kHz. When applied to a commercial submarine equipment with a sensing distance of 900 km, UCAMZI yields a RMSE of approximately 301 m. The UCAMZI has application potential in security monitoring of ultra-long-distance submarine optical cables.

  • Ultra-Long-Distance Interferometric DOFS Over 300 km Without In-Line Repeater

    by Feng Shi, Rui Jin, Yunke Du, Chao Wang, Bo Jia

    An interferometric distributed optical fiber sensing system with a sensing range over 300 km is proposed. This system employs Raman fiber amplifier with bidirectional pumping in conjunction with a bidirectional EDFA, without the use of in-line relay amplifier. Through an analysis of the interference light transmission within the system, the minimum SNR required to effectively demodulate the phase is determined, and the corresponding sensing fiber loss is experimentally quantified. By utilizing ultra-low-loss G654E single mode fiber, the system can achieve high-performance sensing without signal degradation within the 300 km range, and the positioning error is less than 8 m. To the best of our knowledge, this is currently the longest distributed optical fiber sensing system without in-line repeater. The system offers potential applications in fields such as perimeter security and submarine cable monitoring.


  • Position Characteristics Removing Algorithm for Pattern Recognition in Interferometric DOFS

    by Yunke Du, Rui Jin, Churui Li, Bo Ma, Yuzhe Sun, Chenyue He, Yang Yan, Chao Wang, Bo Jia

    In distributed optical fiber sensing system, pattern recognition is employed to identify vibration signals. This letter introduces a novel algorithm for removing position characteristics to enhance the pattern recognition accuracy based on single core feedback interference system. The detrimental effects of position characteristics on signal identification are analyzed and simulated. Experiments using a hybrid Mach-Zehnder and Sagnac interferometer combined with a convolutional neural network show a classification accuracy of 98.1% across seven types of position-independent vibration signals. This represents a 1.1% improvement over original signals. Focused on accurately classifying vibration signals from various locations, the position characteristics removing algorithm allows classifier trained solely on data collected from a single point, with enhanced accuracy and the ability to generalize across different signal types and locations.