• DocumentCode
    2611693
  • Title

    Fast three-dimensional GPR forward and inverse scattering based on wideband diagonal tensor approximation

  • Author

    Huang, Yueqin ; Liu, Qing H. ; Zhang, Jianzhong

  • Author_Institution
    Sch. of Inf. Sci. & Technol., Xiamen Univ., Xiamen, China
  • fYear
    2010
  • fDate
    9-12 May 2010
  • Firstpage
    1
  • Lastpage
    1
  • Abstract
    A fast three-dimensional (3D) ground penetrating radar (GPR) forward and inverse scattering method is presented based on the diagonal tensor approximation (DTA). For the forward scattering problem, the first iteration about the frequency-dependent scattering coefficients in the DTA is used. By further evaluating the scattering tensor at the GPR center frequency to remove its frequency dependence, the DTA is simplified while maintaining higher accuracy than the classical Born approximation method. For the inverse scattering problem, the DTA-based algorithm is derived for the plane-wave GPR, and can be implemented efficiently by the NUFFT combined with the stabilized bi-conjugate gradient FFT (BiCGS-FFT) method. The proposed inversion algorithm is much cheaper than the existing DTA-based methods, and has a higher accuracy and a wider range of applicability than the fast method based on the Born approximation.
  • Keywords
    fast Fourier transforms; gradient methods; ground penetrating radar; scattering; tensors; Born approximation method; DTA-based algorithm; GPR center frequency; bi-conjugate gradient FFT method; forward scattering method; frequency-dependent scattering coefficients; inverse scattering method; three-dimensional ground penetrating radar; wideband diagonal tensor approximation; Approximation algorithms; Approximation methods; Electric variables measurement; Frequency dependence; Ground penetrating radar; Information science; Inverse problems; Radar scattering; Tensile stress; Wideband;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electromagnetic Field Computation (CEFC), 2010 14th Biennial IEEE Conference on
  • Conference_Location
    Chicago, IL
  • Print_ISBN
    978-1-4244-7059-4
  • Type

    conf

  • DOI
    10.1109/CEFC.2010.5481592
  • Filename
    5481592