Scholarlink

<p>Distributed acoustic sensing (DAS) offers unprecedented spatial resolution and long-range coverage for rockfall hazard monitoring. However, in acoustically complex field environments, strong ambient wavefields heavily mask transient rockfall impact signatures. This interference hinders the accurate extraction of signal onset times, severely degrading spatial localization accuracy. To overcome this operational bottleneck, this study proposes a high-precision rockfall localization framework that integrates adaptive signal isolation with a time-difference-of-arrival algorithm. Using DAS data acquired from controlled artificial rockfall experiments, we systematically evaluated the signal-enhancement performance of several advanced modal decomposition and wavelet-based techniques to address the non-stationary characteristics of DAS recordings. Results demonstrate that an optimized variational mode decomposition approach maximizes kinematic signal fidelity. These isolated rockfall signatures achieved an average signal-to-noise ratio of up to 26.94&#xa0;dB, representing a 58% improvement over unprocessed recordings and outperforming alternative isolation frameworks by 12%–30%. Leveraging the precise arrival times extracted from the enhanced signatures, a station-pair time-difference-of-arrival approach successfully localized the primary impact sources, reducing the average spatial localization error to just 4.5&#xa0;m—less than half the error (&gt; 10&#xa0;m) yielded by conventional methods. Furthermore, spatial error analysis demonstrates that while parallel array configurations resolve pseudo-source ambiguities along the fiber axis, deploying nonlinear or orthogonal cable layouts is essential to fully eliminate perpendicular spatial uncertainties. Ultimately, this study validates a highly effective, integrated approach for mitigating complex ambient wavefields, providing a robust operational framework for the real-time, high-precision spatial tracking of geological hazards in rugged mountain topographies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

High-precision rockfall source localization in acoustically complex environments using distributed acoustic sensing

  • 期刊论文

  • 发表日期：

    2026年6月30日

查看全文
  • Fei-Long Zhang,
  • Cheng-Cheng Zhang,
  • Zheng Wang,
  • Tao Xie,
  • Bin Shi

摘要

Distributed acoustic sensing (DAS) offers unprecedented spatial resolution and long-range coverage for rockfall hazard monitoring. However, in acoustically complex field environments, strong ambient wavefields heavily mask transient rockfall impact signatures. This interference hinders the accurate extraction of signal onset times, severely degrading spatial localization accuracy. To overcome this operational bottleneck, this study proposes a high-precision rockfall localization framework that integrates adaptive signal isolation with a time-difference-of-arrival algorithm. Using DAS data acquired from controlled artificial rockfall experiments, we systematically evaluated the signal-enhancement performance of several advanced modal decomposition and wavelet-based techniques to address the non-stationary characteristics of DAS recordings. Results demonstrate that an optimized variational mode decomposition approach maximizes kinematic signal fidelity. These isolated rockfall signatures achieved an average signal-to-noise ratio of up to 26.94 dB, representing a 58% improvement over unprocessed recordings and outperforming alternative isolation frameworks by 12%–30%. Leveraging the precise arrival times extracted from the enhanced signatures, a station-pair time-difference-of-arrival approach successfully localized the primary impact sources, reducing the average spatial localization error to just 4.5 m—less than half the error (> 10 m) yielded by conventional methods. Furthermore, spatial error analysis demonstrates that while parallel array configurations resolve pseudo-source ambiguities along the fiber axis, deploying nonlinear or orthogonal cable layouts is essential to fully eliminate perpendicular spatial uncertainties. Ultimately, this study validates a highly effective, integrated approach for mitigating complex ambient wavefields, providing a robust operational framework for the real-time, high-precision spatial tracking of geological hazards in rugged mountain topographies.