RTM using effective boundary saving: A staggered grid GPU implementation [pdf 1.8M] Pengliang Yang, Baoli Wang, and Jinghuai Gao GPU has become a booming technology in reverse time migration (RTM) to perform the intensive computation. Compared with saving forward modeled wavefield on the disk, RTM via wavefield reconstruction using saved boundaries on device is a more efficient method because computation is much faster than CPU-GPU data transfer.
In this paper, we introduce the effective boundary saving strategy in backward reconstruction for RTM. The minimum storage requirement for regular and staggered grid finite difference is determined for perfect reconstruction of the source wavefield. Particularly, we implement RTM using GPU programming, combining staggered finite difference scheme with convolutional perfectly matched layer (CPML) boundary condition. We demonstrate the validity of the proposed approach and CUDA codes with numerical example and imaging of benchmark models.
A graphics processing unit implementation of time-domain full-waveform inversion [pdf 1008K] Pengliang Yang, Jinghuai Gao, and Baoli Wang The graphics processing unit (GPU) has become a popular device for seismic imaging and inversion due to its superior speedup performance. In this paper we implement GPU-based full waveform inversion (FWI) using the wavefield reconstruction strategy. Because the computation on GPU is much faster than CPU-GPU data communication, in our implementation the boundaries of the forward modeling are saved on the device to avert the issue of data transfer between host and device. The Clayton-Enquist absorbing boundary is adopted to maintain the efficiency of GPU computation. A hybrid nonlinear conjugate gradient algorithm combined with the parallel reduction scheme is utilized to do computation in GPU blocks. The numerical results confirm the validity of our implementation.
Seislet-based morphological component analysis using scale-dependent exponential shrinkage [pdf 460K] Pengliang Yang and Sergey Fomel Morphological component analysis (MCA) is a powerful tool used in image processing to separate different geometrical components (cartoons and textures, curves and points etc). MCA is based on the observation that many complex signals may not be sparsely represented using only one dictionary/transform, however can have sparse representation by combining several over-complete dictionaries/transforms. In this paper we propose seislet-based MCA for seismic data processing. MCA algorithm is reformulated in the shaping-regularization framework. Successful seislet-based MCA depends on reliable slope estimation of seismic events, which is done by plane-wave destruction (PWD) filters. An exponential shrinkage operator unifies many existing thresholding operators and is adopted in scale-dependent shaping regularization to promote sparsity. Numerical examples demonstrate a superior performance of the proposed exponential shrinkage operator and the potential of seislet-based MCA in application to trace interpolation and multiple removal.
A numerical tour of wave propagation [pdf 2.6M] Pengliang Yang This tutorial is written for beginners as an introduction to basic wave propagation using finite difference method, from acoustic and elastic wave modeling, to reverse time migration and full waveform inversion. Most of the theoretical delineations summarized in this tutorial have been implemented in Madagascar with Matlab, C and CUDA programming, which will benefit readers' further study.
Fourier pseudo spectral method for attenuative simulation with fractional Laplacian [pdf 116K] Pengliang Yang This tutorial is devoted to pseudo-spectral method (PSM) by reorganizing the course material from Prof. Heiner Igel and a paper by J.M. Carcione, to illustrate how to implement viscoacoustic wave simulation based on fractional Laplacian operator.
From modeling to full waveform inversion: A hands-on tour using Madagascar [pdf 864K] Pengliang Yang This tutorial is devoted to Madagascar school 2016 Zurich. In this tutorial, there are two aspects we would like to explore:
Madagascar functionality, which is the tool. We may consider Madagascar to facilitate our research, from the numerical test to publication.
Scientific aspects, which are the key things we care. Even though we are playing a game with simple exercise, we have to think about the scientific enhancement/improvement to polish the techniques used in modeling and inversion applications.