• DocumentCode
    1498013
  • Title

    Electromagnetic Scattering of Randomly Rough Soil Surfaces Based on Numerical Solutions of Maxwell Equations in Three-Dimensional Simulations Using a Hybrid UV/PBTG/SMCG Method

  • Author

    Huang, Shaowu ; Tsang, Leung

  • Author_Institution
    Dept. of Electr. Eng., Univ. of Washington, Seattle, WA, USA
  • Volume
    50
  • Issue
    10
  • fYear
    2012
  • Firstpage
    4025
  • Lastpage
    4035
  • Abstract
    A hybrid UV/PBTG/SMCG method is developed to accelerate the solution of NMM3D for 3-D electromagnetic wave scattering by random rough soil surfaces. It takes only 1.5 min using 16 processors on a cluster of NSF TeraGrid to compute 3-D solution of Maxwell equations of 8 by 8 square wavelengths. With the improved computational efficiency, we computed results using areas up to 32 by 32 square wavelengths. With the larger surface area, we were able to compute cases with larger rms heights up to 8 cm at SMAP radar frequency of 1.26 GHz that corresponds to kh = 2.28. New results computed include the cross-polarization. The five tests on the accuracy of results were performed: convergence with realizations, convergence with discrete samplings, convergence with sample surface sizes, energy conservation for each realization, and reciprocity for each realization. The numerical results show that for a single realization VH = HV within 0.5 dB in the backscattering direction showing that reciprocity is obeyed. Results were compared with experimental data, and both co-polarization and cross-polarization were in good agreements with experimental data.
  • Keywords
    Maxwell equations; convergence of numerical methods; electromagnetic wave scattering; geophysics computing; grid computing; radar theory; radiowave propagation; remote sensing by radar; soil; 3D Maxwell equation solution; 3D electromagnetic wave scattering; 3D simulations; NMM3D solution accleration; NSF TeraGrid; SMAP radar frequency; cross polarization; discrete sampling convergence; electromagnetic scattering; frequency 1.26 GHz; hybrid UV-PBTG-SMCG method; randomly rough soil surfaces; realization convergence; realization energy conservation; realization reciprocity; sample surface size convergence; Chebyshev approximation; Integrated circuits; Interpolation; Rough surfaces; Surface impedance; Surface roughness; Surface waves; Microwave remote sensing; SMCG; random rough surface; soil moisture;
  • fLanguage
    English
  • Journal_Title
    Geoscience and Remote Sensing, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0196-2892
  • Type

    jour

  • DOI
    10.1109/TGRS.2012.2189776
  • Filename
    6185659