Multi-synchrosqueezing polynomial chirplet transform and its application to seismic thin interbeds analysis

Abstract

Thin interbeds are important geological structures in seismic exploration, and the analysis of their thickness variations is a key step in seismic interpretation. As a typical signal-processing technology, the time–frequency analysis method maps one-dimensional signals to the two-dimensional time–frequency domain, which can capture the changes of the instantaneous frequency. On this basis, the time–frequency analysis method can be used to analyse the thickness changes of thin interbeds. To analyse the thickness changes more accurately, the adopted time–frequency analysis method needs to have high time–frequency resolution and robustness. This paper proposes a novel method named the multi-synchrosqueezing polynomial chirplet transform. First, the second-order instantaneous frequency estimation operator is obtained through the corrected polynomial chirplet transform. Then, through squeezing and rearranging, the fuzzy time–frequency energy is gradually concentrated on the corresponding second-order multiple instantaneous frequency estimation operator to obtain the multi-synchrosqueezing polynomial chirplet transform. Simulated signals are used to demonstrate that multi-synchrosqueezing polynomial chirplet transform has robustness while improving time–frequency resolution. Simultaneously, simulated and real seismic signals are used to verify that the proposed method can analyse the thickness variation of thin interbeds.