Full text loading...
-
A Study of Different Greedy Algorithms for Crossline Wavefield Reconstruction Using Multimeasurement Towed Marine Data
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 76th EAGE Conference and Exhibition 2014, Jun 2014, Volume 2014, p.1 - 5
Abstract
A general framework for the joint interpolation and 3D deghosting of severely aliased data using multimeasurement data is provided. This is used to solve the (severely) underdetermined inverse problem of reconstructing the upgoing pressure data in the crossline wavenumber domain. This work aims to study the difference between different greedy algorithms to solve the inverse problem. In particular, four algorithms are discussed - Generalized Matching Pursuit (GMP), Orthogonal GMP (OGMP), GMP with a final orthogonalization step (GMPr), and Optimized Orthogonal GMP (OOGMP). All these algorithms iteratively estimate the spatial wavenumbers and the complex amplitudes associated with them. However, they have different rates of convergence. We highlight the main differences between these algorithms in terms of complexity and performance