Phase sensitive optical time domain reflectometry (Φ-OTDR) has been widely used in the precise acquisition of the vibration signal along the sensing optical fiber, and has attracted more and more interest. However, polarization fading and coherent fading usually lead to the phase hopping of the recovered signal, thus it is difficult to realize the continuous measurement of vibration, especially the low frequency vibration. The polarization diversity and multi-frequency detection schemes which are usually used to solve the above problems need additional hardware, thus increasing the cost and complexity of the system. In this article, we propose a statistical method based on nearest neighbor analysis. For each interrogation fiber separation, the phases of the sub-separations are evaluated and selected to reconstruct the vibration. In the experiment, the vibrations with frequencies as low as 0.2 Hz are loaded on the fiber and recovered. The results show that the method can effectively suppress the phase hopping, correct distorted vibration and improve the performance of low frequency vibration measurement of Φ-OTDR system on the basis of only numerical calculation.

中文翻译：

---

基于最近邻分析的 OTDR 中的衰落噪声抑制

相敏光时域反射计（Φ-OTDR）已广泛应用于沿传感光纤的振动信号的精确采集，并越来越引起人们的兴趣。然而，极化衰落和相干衰落通常会导致恢复信号的相位跳跃，从而难以实现对振动，尤其是低频振动的连续测量。通常用于解决上述问题的极化分集和多频检测方案需要额外的硬件，从而增加了系统的成本和复杂度。在本文中，我们提出了一种基于最近邻分析的统计方法。对于每个询问纤维分离，子分离的相位被评估和选择以重建振动。在实验中，频率低至 0.2 Hz 的振动被加载到光纤上并恢复。结果表明，该方法仅通过数值计算就能有效抑制相位跳变、校正畸变振动，提高Φ-OTDR系统低频振动测量的性能。