Title :
Multidomain pseudospectral time-domain simulations of scattering by objects buried in lossy media
Author :
Fan, Guo-Xin ; Liu, Qing Huo ; Hesthaven, Jan S.
Author_Institution :
Dept. of Electr. & Comput. Eng., Duke Univ., Durham, NC, USA
fDate :
6/1/2002 12:00:00 AM
Abstract :
A multidomain pseudospectral time-domain (PSTD) method with a newly developed well-posed PML is introduced as an accurate and flexible tool for the modeling of electromagnetic scattering by 2-D objects buried in an inhomogeneous lossy medium. Compared with the previous single-domain Fourier PSTD method, this approach allows for an accurate treatment of curved geometries with subdomains, curvilinear mapping, and high-order Chebyshev polynomials. The effectiveness of the algorithm is confirmed by an excellent agreement between the numerical results and analytical solutions for perfectly conducting as well as permeable dielectric cylinders. The algorithm has been applied to model various ground-penetrating radar (GPR) applications involving curved objects in a lossy half space with an undulating surface. This multidomain PSTD algorithm is potentially a very useful tool for simulating antennas near complex objects and inhomogeneous media.
Keywords :
backscatter; buried object detection; finite difference time-domain analysis; geophysical techniques; radar cross-sections; radar theory; remote sensing by radar; terrain mapping; 2-D objects; Chebyshev polynomials; algorithm; buried object detection; curved geometry; curvilinear mapping; electromagnetic scattering; finite-difference method; geophysical measurement technique; ground penetrating radar; inhomogeneous medium; land surface; lossy media; multidomain pseudospectral time domain simulation; radar remote sensing; radar scattering; subdomain; terrain mapping; Algorithm design and analysis; Chebyshev approximation; Dielectrics; Electromagnetic modeling; Electromagnetic scattering; Geometry; Ground penetrating radar; Polynomials; Radar scattering; Time domain analysis;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
DOI :
10.1109/TGRS.2002.800272