An Enhanced Hybrid Method for Solving Inverse Scattering Problems

A new hybrid method for solving inverse scattering problems in time domain is proposed. In this method, a two-step optimization-based routine is used in which direct and so called expansion methods are utilized simultaneously to obtain more precise and rapid reconstruction. In the first step, a coar...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on magnetics 2009-03, Vol.45 (3), p.1534-1537
Main Authors: Semnani, A., Kamyab, M.
Format: Article
Language:English
Subjects:
Algorithms
Conductivity
Convergence
Cross-disciplinary physics: materials science
rheology
Exact sciences and technology
Finite difference methods
Finite-difference time-domain (FDTD)
Fourier series
Inverse problems
Inverse scattering
Magnetism
Materials science
Mathematical analysis
Optimization
Optimization methods
Other topics in materials science
Particle swarm optimization
particle swarm optimizer (PSO)
Permittivity
Physics
Reconstruction
Routines
Solvers
Studies
Time domain
Time domain analysis
Time measurement
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A new hybrid method for solving inverse scattering problems in time domain is proposed. In this method, a two-step optimization-based routine is used in which direct and so called expansion methods are utilized simultaneously to obtain more precise and rapid reconstruction. In the first step, a coarse solution is achieved with the help of truncated cosine Fourier series expansion method. Then, with this solution as an initial guess for the second step, direct optimization is used to obtain a much more accurate solution of the problem. In both steps, finite difference time domain and particle swarm optimization are used as an appropriate electromagnetic solver and global optimization routine, respectively. The most important advantage of this method is that because of a suitable initial answer in direct optimization routine, sensitivity of the algorithm to the regularization parameter is decreased and convergence of the results is completely guaranteed. Performance of the proposed method is demonstrated and compared with expansion and conventional direct optimization methods for several 1-D permittivity and conductivity profile reconstruction cases.
ISSN:0018-9464
1941-0069
DOI:10.1109/TMAG.2009.2012735