Weighted average instantaneous frequency extraction via time-frequency analysis

Instantaneous frequency (IF) extracted from seismic data is generally used in describing time-frequency features of nonstationary seismic data. However, the conventional extraction methods, such as the IF extraction based on the Hilbert transform (HT), are seriously sensitive to noise and spikes, which brings difficulty for seismic attributes analysis and interpretation. By introducing a small window, weighted average instantaneous frequency (WAIF) can reduce frequency spikes well and improve interpretability. In this paper, we propose a novel WAIF extraction method based on time-frequency analysis, which is called synchrosqueezing transform taking the three parameter wavelet (SST-TPW). We introduce this promising WAIF extraction method to seismic data processing and show its effectiveness. Results on synthetic signals and field data demonstrate the excellent performance of the proposed method.