Generalized Cross-Correlation Delay Estimation Algorithm Based on Improved LMS Adaptive Filtering

The Phase-Sensitive Optical Time Domain Reflectometer
$(\Phi-\text{OTDR})$
Distributed Acoustic Sensing (DAS) system is a pioneering technological breakthrough with profound implications in industrial, military, security, and environmental monitoring domains. This system employs optical fiber sensing technology to monitor the transmission of sound waves in optical fiber, enabling precise sound source localization and distributed monitoring. This, in turn, offers critical data and insights for various applications. Addressing the time delay estimation challenge in DAS, researchers are exploring a more accurate and efficient alternative to the commonly used generalized cross-correlation method. Although widely adopted, the current method has limitations, primarily related to the varying influence of different weighting coefficients on time delay estimation, leading to potential accuracy and stability issues. To address this, a new delay error estimation method is proposed, specifically designed for the DAS sound source location system. This strategy incorporates a dynamic step minimum mean square error adaptive filter to effectively manage the compromise between the rate of convergence and constant error of the minimum mean square error (LMS) algorithm. The key concept revolves around dynamically adapting the step size based on the prevailing conditions to achieve more precise delay estimation results. We have done a lot of simulation experiments with MATLAB, compared with the traditional generalized cross-correlation method, and evaluated the effectiveness of the new algorithm. The experimental results show that our method has a great improvement in time delay estimation, and the proposed method effectively solves the difficult problem of weight coefficient selection in the traditional method, thus improving the accuracy and robustness of the sound source location system.