A methodology of adaptive time series analysis, based on Empirical Mode Decomposition (EMD), and on its time varying version tvf-EMD has been applied to strain data from the gravitational wave interferometer (IFO) Virgo in order to characterise scattered light noise affecting the sensitivity of the IFO in the detection frequency band. Data taken both during hardware injections, when a part of the IFO is put in oscillation for detector characterisation purposes, and during periods of science mode, when the IFO is fully locked and data are used for the detection of gravitational waves, were analysed. The adaptive nature of the EMD and tvf-EMD algorithms allows them to deal with nonlinear non-stationary data and hence they are particularly suited to characterise scattered light noise which is an intrinsically nonlinear and non-stationary noise. Obtained results show that tvf-EMD algorithm allows to obtain more precise results compared to the EMD algorithm, yielding higher cross-correlation values with the auxiliary channels that are the culprits of scattered light noise.

中文翻译：

---

使用 tvf-EMD 自适应算法在 Virgo 干涉仪上表征散射光噪声

一种基于经验模式分解 (EMD) 及其时变版本 tvf-EMD 的自适应时间序列分析方法已应用于来自引力波干涉仪 (IFO) Virgo 的应变数据，以表征影响IFO 在检测频带中的灵敏度。分析了在硬件注入期间（将 IFO 的一部分置于振荡以用于探测器表征目的）以及在科学模式期间（当 IFO 完全锁定且数据用于检测引力波时）获取的数据。EMD 和 tvf-EMD 算法的自适应特性允许它们处理非线性非平稳数据，因此它们特别适合表征散射光噪声，这是一种本质上非线性和非平稳噪声。