DAS coupling noise suppression using Wavelet and DCT dictionary based on sparse optimization

The distributed acoustic sensing (DAS) system, which uses an optical fiber cable in vertical seismic profiling (VSP) data acquisition, has been launched as one promising technology. The acquired high quality data from DAS can produce high-precision VSP images, obtain more detailed check shots. Compared with common data acquisition, the cost of the repeated data acquisition with DAS is much lower. However, the data from DAS acquisition suffers from strong coherent noise which is caused by the physical placement of the wireline in the well. In this work, we analyze the morphological characteristics of DAS coupling noise and try to propose one DAS coupling noise suppressing method based on sparse optimization. We use continuous wavelet transform and discrete cosine transform to represent the desired signals and coupling noise sparsely. We combine these two dictionaries to form an over-complete dictionary and introduce the morphological component analysis to remove DAS coupling noise. The proposed method is applied to one synthetic dataset and one field dataset to validate its effectiveness.