Combining synchrosqueezed wavelet transforms and empirical mode decomposition for filtering seismic random noise

Compared with continuous wavelet transform (CWT), syn-chrosqueezing transform (SST) decompose a signal with higher precision in time and frequency, which can be utilized to recognize and reduce random noise. Considering the mix- ture of noise and effective components in time-frequency map, the problem is that we simply neglect components of low en- ergy leads to bad preservation for signal amplitude. We adopt empirical mode decomposition (EMD) to improve the SST re- sults. We propose a new method that utilize the decomposition characteristic of EMD which decomposes a signal to several modes from high to low frequency and to take advantage of the time-frequency filtering characteristic of SST which can recognize the valid signal component in time-frequency map in order to achieve effective random noise reduction together with good amplitude preservation. Numerical experiments on synthetic and real seismic data show its effectiveness.