Transient Electromagnetic Noise Suppression Based on MVMD and GMDFA

Noise suppression is an important topic in transient electromagnetic (TEM) data processing and interpretation. TEM data is a typical broadband signal, which makes it difficult to separate the signal in the whole frequency band. The conventional methods tend to process data trace by trace, ignoring the connections between channels. This paper propose a workflow based on multichannel variational mode decomposition (MVMD) and genetic multichannel detrended fluctuation analysis (GMDFA) to deal with the noise in multichannel TEM data. The proposed workflow first decomposes the 2-D TEM data into several intrinsic mode functions (IMFs) using MVMD. Then we use GMDFA to screen out the signal IMFs and remove the noise traces in signal IFMs using principal component analysis (PCA). Finally, the signal IMFs are summed up to reconstruct TEM signal. Field result shows that the proposed workflow effectively suppresses noise compared with traditional denoising methods.