A hybrid method for noise suppression using variational mode decomposition and singular spectrum analysis

Abstract

Seismic noise suppression is an important step in the seismic imaging community. We propose a variational mode decomposition (VMD) based method to attenuate spatially incoherent random noise. The VMD method has a solid theoretical foundation of mathematics and high calculation efficiency. Besides, compared with the recursive mode decomposition algorithms, e.g., the EMD and EEMD methods, it has advantages in solving the mode mixing problem and more powerful anti-noise performance. The VMD method can adaptively decompose a seismic signal into several intrinsic mode functions (IMF). Removing the highly oscillating IMF components can suppress the high frequency/wavenumber noise. However, the VMD algorithm has a drawback that the mode number needs to be set artificially before decomposition. The mode number is difficult to estimate when a signal is a complex nonlinear and non-stationary one. Thus, the signal is still coupled in low frequency/wavenumber noise. Thereupon, the singular spectrum analysis (SSA) algorithm is applied to filter the low-frequency/wavenumber noise. In this paper, a simulated seismic dataset and a real seismic dataset are analyzed by the proposed algorithm. The results indicate that the proposed algorithm is robust to noise and has high de-noising precision.