High-Resolution Time-Frequency Analysis of Seismic Signals Using Synchrosqueezed Wavelet Transform

Abstract

Seismic signals are inherently nonstationary and often contain complex frequency components that are challenging to characterize using conventional time–frequency analysis methods. To overcome the limitations of techniques such as the Short-Time Fourier Transform (STFT) and even its synchrosqueezed variants (FSST and FSST2), this study introduces the Synchrosqueezed Wavelet Transform (SSWT) for enhanced time– frequency representation. By combining the multiscale analysis capability of wavelet transforms with adaptive spectral squeezing, SSWT achieves superior resolution and energy concentration. Experiments on both synthetic and real seismic data demonstrate that SSWT effectively resolves rapid frequency variations and improves the interpretability of instantaneous frequency features, offering significant potential for high-precision seismic reservoir characterization