Time-frequency analysis of seismic data based on Basis Pursuit

Seismic time-frequency analysis is an important seismic data processing method which could indicate rock properties and achieve the purpose of exploration via analyzing the time information and frequency information carried by seismic data. Wigner-Ville Distribution (WVD) is a kind of typical bilinear time-frequency analysis method, but there are serious crossing interference items when it is applied to multi-component signal. Basis Pursuit (BP) is an effective algorithm in the area of sparse decomposition owing to its flexibility and self-adaption. This paper presents a crossing items suppression method based on BP. Basis Pursuit is applied to complete the sparse decomposition and the seismic trace may be decomposed into a series of wavelets that match their time-frequency signature. Then this paper combines the consequence of decomposition with WVD for the time-frequency analysis of seismic data. The simulation results show that the new method based on Basis Pursuit not only reduces the crossing items caused by WVD but also improves the time-frequency resolution and the accuracy of exploration.