A preconditioned inexact newton method for nonlinear sparse electromagnetic imaging

Handle URI:
http://hdl.handle.net/10754/564070
Title:
A preconditioned inexact newton method for nonlinear sparse electromagnetic imaging
Authors:
Desmal, Abdulla ( 0000-0003-0861-8908 ) ; Bagci, Hakan ( 0000-0003-3867-5786 )
Abstract:
A nonlinear inversion scheme for the electromagnetic microwave imaging of domains with sparse content is proposed. Scattering equations are constructed using a contrast-source (CS) formulation. The proposed method uses an inexact Newton (IN) scheme to tackle the nonlinearity of these equations. At every IN iteration, a system of equations, which involves the Frechet derivative (FD) matrix of the CS operator, is solved for the IN step. A sparsity constraint is enforced on the solution via thresholded Landweber iterations, and the convergence is significantly increased using a preconditioner that levels the FD matrix's singular values associated with contrast and equivalent currents. To increase the accuracy, the weight of the regularization's penalty term is reduced during the IN iterations consistently with the scheme's quadratic convergence. At the end of each IN iteration, an additional thresholding, which removes small 'ripples' that are produced by the IN step, is applied to maintain the solution's sparsity. Numerical results demonstrate the applicability of the proposed method in recovering sparse and discontinuous dielectric profiles with high contrast values.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Center for Uncertainty Quantification in Computational Science and Engineering (SRI-UQ); Electrical Engineering Program; Computational Electromagnetics Laboratory
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
IEEE Geoscience and Remote Sensing Letters
Issue Date:
Mar-2015
DOI:
10.1109/LGRS.2014.2349935
Type:
Article
ISSN:
1545598X
Appears in Collections:
Articles; Electrical Engineering Program; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorDesmal, Abdullaen
dc.contributor.authorBagci, Hakanen
dc.date.accessioned2015-08-03T12:30:44Zen
dc.date.available2015-08-03T12:30:44Zen
dc.date.issued2015-03en
dc.identifier.issn1545598Xen
dc.identifier.doi10.1109/LGRS.2014.2349935en
dc.identifier.urihttp://hdl.handle.net/10754/564070en
dc.description.abstractA nonlinear inversion scheme for the electromagnetic microwave imaging of domains with sparse content is proposed. Scattering equations are constructed using a contrast-source (CS) formulation. The proposed method uses an inexact Newton (IN) scheme to tackle the nonlinearity of these equations. At every IN iteration, a system of equations, which involves the Frechet derivative (FD) matrix of the CS operator, is solved for the IN step. A sparsity constraint is enforced on the solution via thresholded Landweber iterations, and the convergence is significantly increased using a preconditioner that levels the FD matrix's singular values associated with contrast and equivalent currents. To increase the accuracy, the weight of the regularization's penalty term is reduced during the IN iterations consistently with the scheme's quadratic convergence. At the end of each IN iteration, an additional thresholding, which removes small 'ripples' that are produced by the IN step, is applied to maintain the solution's sparsity. Numerical results demonstrate the applicability of the proposed method in recovering sparse and discontinuous dielectric profiles with high contrast values.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.subjectElectromagnetic (EM) imagingen
dc.subjectinexact Newton (IN)en
dc.subjectsparse optimizationen
dc.subjectthresholded Landweber (LW)en
dc.titleA preconditioned inexact newton method for nonlinear sparse electromagnetic imagingen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentCenter for Uncertainty Quantification in Computational Science and Engineering (SRI-UQ)en
dc.contributor.departmentElectrical Engineering Programen
dc.contributor.departmentComputational Electromagnetics Laboratoryen
dc.identifier.journalIEEE Geoscience and Remote Sensing Lettersen
kaust.authorBagci, Hakanen
kaust.authorDesmal, Abdullaen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.