Dreamlet based Gaussian packet migration
Yu Geng, Ru‐Shan Wu, Jinghuai Gao · 2014
Gaussian packets are high-frequency asymptotic solutions of the wave equation. Gaussian packets have both Gaussian shaped time-frequency localization and space-direction localization. Its evolution can be obtained by ray tracing and dynamic ray equations. Through investigating into the relationship between Gaussian packets and Dreamlet atoms, we proposed and tested an effective decomposition method of seismic data (surface reflection profiling) into Gaussian packets. We test the poststack depth migration to the datasets of a 4-layer model and the 2D SEG-EAGE model to demonstrate the method's validity and accuracy. From the results, we see that the Dreamlet based Gaussian packet migration can obtain the correct main features of the subsurface structures with properly smoothed velocity model. This shows the potential of Gaussian packet migration as the fast method of preliminary investigation.