Principal component analysis in the wavelet domain

作者：

Highlights：

• Dimension reduction of multiple non-stationary time series and for identifying important features.

• Compare time-varying spectra and the time-scale-dependent spectra.

• Detect the unusual pattern of the data in the seismic wave.

摘要

•Dimension reduction of multiple non-stationary time series and for identifying important features.•Compare time-varying spectra and the time-scale-dependent spectra.•Detect the unusual pattern of the data in the seismic wave.

论文关键词：Principal component analysis,Non-stationary time series,Wavelet process,Feature extraction,Seismic data

论文评审过程：Received 28 October 2020, Revised 30 April 2021, Accepted 3 June 2021, Available online 10 June 2021, Version of Record 23 June 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108096