Noise suppression of distributed acoustic sensing vertical seismic profile data based on time–frequency analysis.

Distributed acoustic sensing (DAS) technology is a novel technology applied in vertical seismic profile (VSP) exploration, which has many advantages, such as low cost, high precision, strong tolerance to harsh acquisition environment. However, the field DAS-VSP data are often disturbed by complex background noise and coupling noise with strong energy, affecting the quality of seismic data seriously. Therefore, we develop a time–frequency analysis method based on low-rank and sparse matrix decomposition (LSMD) and data position points distribution maps (DPM) to separate signals from noise. We adopt Multisynchrosqueezing Transform to construct the approximate ideal time–frequency representation of DAS data, which reduces the difficulty of signal to noise separation and avoids the loss of some effective information to a certain extent. The LSMD is performed to separate the signal component and noise component preliminarily. In addition, combined with the separated low-rank matrix and sparse matrix, we propose the DPM to improve the accuracy of signal component extraction and the recovery ability of the method for weak signals through the joint analysis of the maps in time domain and frequency domain. Both synthetic and field experiments show that the proposed method can suppress coupling noise and background noise and recover weak energy signals in DAS VSP data effectively. [ABSTRACT FROM AUTHOR]