Title :
TE Versus TM for the Shape Reconstruction of 2-D PEC Targets Using the Level-Set Algorithm
Author :
Hajihashemi, Mohammad Reza ; El-Shenawee, Magda
Author_Institution :
Dept. of Electr. Eng., Univ. of Arkansas, Fayetteville, AR, USA
fDate :
3/1/2010 12:00:00 AM
Abstract :
The transverse electric (TE) polarization for shape reconstruction of perfect electric conducting 2-D targets is presented. The deformation velocity for the TE polarization case is implemented in the level-set algorithm. A comparison between the reconstruction CPU time between the TE and transverse magnetic (TM) polarizations is discussed. The numerical results show that retrieving the shape and location of multiple targets of arbitrary cross sections becomes computationally intensive when illumination with TE-polarized waves is used. If the orientation of the unknown cylinders is a priori known, the TM-polarized waves provide faster reconstruction results with the same accuracy compared with the TE-polarized waves. Upon corrupting the synthetic data with Gaussian noise up to signal-to-noise ratio of 5 dB, the TM polarization seems to provide more accurate results compared with the TE case.
Keywords :
Gaussian noise; geophysical techniques; geophysics computing; CPU time; Gaussian noise; TE polarization; TE-polarized waves; TM polarization; TM-polarized waves; arbitrary cross sections; computational electromagnetics; deformation velocity; electric conducting 2D targets; level-set algorithm; shape reconstruction; signal-to-noise ratio; transverse electric polarization; transverse magnetic polarization; Computational electromagnetics; inverse scattering; level-set method; shape reconstruction;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
DOI :
10.1109/TGRS.2009.2029698