当前位置: X-MOL 学术SIAM J. Appl. Dyn. Syst. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Time Series Source Separation Using Dynamic Mode Decomposition
SIAM Journal on Applied Dynamical Systems ( IF 2.1 ) Pub Date : 2020-05-11 , DOI: 10.1137/19m1273256
Arvind Prasadan , Raj Rao Nadakuditi

SIAM Journal on Applied Dynamical Systems, Volume 19, Issue 2, Page 1160-1199, January 2020.
The dynamic mode decomposition (DMD) extracted dynamic modes are the nonorthogonal eigenvectors of the matrix that best approximates the one-step temporal evolution of the multivariate samples. In the context of dynamical system analysis, the extracted dynamic modes are a generalization of global stability modes. We apply DMD to a data matrix whose rows are linearly independent, additive mixtures of latent time series. We show that when the latent time series are uncorrelated at a lag of one time-step then, in the large sample limit, the recovered dynamic modes will approximate, up to a columnwise normalization, the columns of the mixing matrix. Thus, DMD is a time series blind source separation algorithm in disguise but is different from closely related second-order algorithms such as the second-order blind identification (SOBI) method and the algorithm for multiple unknown signals extraction (AMUSE). All can unmix mixed stationary, ergodic Gaussian time series in a way that kurtosis-based independent components analysis fundamentally cannot. We use our insights on single-lag DMD to develop a higher lag extension, analyze the finite sample performance with and without randomly missing data, and identify settings where the higher lag variant can outperform the conventional single-lag variant. We validate our results with numerical simulations and highlight how DMD can be used in changepoint detection.


中文翻译:

使用动态模式分解的时间序列源分离

SIAM应用动力系统杂志,第19卷,第2期,第1160-1199页,2020年1月。
动态模式分解(DMD)提取的动态模式是矩阵的非正交特征向量,它最近似于多元样本的一步式时间演化。在动力学系统分析的上下文中,提取的动力学模式是全局稳定模式的概括。我们将DMD应用于其行是线性独立的,潜在时间序列的累加混合的数据矩阵。我们表明,当潜伏时间序列在一个时间步长的滞后之间不相关时,则在大样本限制下,恢复的动态模式将近似混合矩阵的列,直至逐列归一化。从而,DMD是一种变相的时间序列盲源分离算法,但与紧密相关的二阶算法(如二阶盲识别(SOBI)方法和用于多个未知信号提取的算法(AMUSE))不同。所有人都可以以基于峰度的独立分量分析从根本上无法做到的方式,解开混合的平稳的,遍历高斯的时间序列。我们利用对单滞后DMD的见识来开发更高的滞后扩展,分析有无随机缺失数据的有限样本性能,并确定高滞后变量可以胜过传统单滞后变量的设置。我们用数值模拟验证了我们的结果,并强调了DMD如何用于变化点检测。遍历的高斯时间序列,这是基于峰度的独立成分分析从根本上无法实现的。我们利用对单滞后DMD的见识来开发更高的滞后扩展,分析有无随机缺失数据的有限样本性能,并确定较高滞后变量可以胜过传统单滞后变量的设置。我们用数值模拟验证了我们的结果,并强调了DMD如何用于变化点检测。遍历的高斯时间序列,这是基于峰度的独立成分分析从根本上无法实现的。我们利用对单滞后DMD的见识来开发更高的滞后扩展,分析有无随机缺失数据的有限样本性能,并确定高滞后变量可以胜过传统单滞后变量的设置。我们用数值模拟验证了我们的结果,并强调了DMD如何用于变化点检测。
更新日期:2020-06-30
down
wechat
bug