• DocumentCode
    345768
  • Title

    Quadratic and linear filters for radio channel prediction

  • Author

    Ekman, Torbjörn ; Kubin, Gernot ; Sternad, Mikael ; Ahlén, Anders

  • Author_Institution
    Dept. of Signals & Syst., Uppsala Univ., Sweden
  • Volume
    1
  • fYear
    1999
  • fDate
    1999
  • Firstpage
    146
  • Abstract
    Using a simple channel model for the multipath mobile radio channel a nonlinear approach to long-range prediction of fading radio channels is motivated in a scenario with scatterers close to the mobile station. The performance of linear and quadratic predictors is evaluated on wideband (6.4 MHz) measurements in the 1800 MHz band. The predictors are applied on estimated complex channel taps to find practical measures on the achievable prediction gain for different prediction horizons. Quadratic predictors, in the form of quadratic Volterra filters, is found to have good modeling properties, but very poor generalization properties as compared to linear predictors
  • Keywords
    UHF filters; UHF radio propagation; electromagnetic wave scattering; fading channels; filtering theory; land mobile radio; multipath channels; prediction theory; 1800 MHz; 6.4 MHz; UHF; channel model; estimated complex channel taps; fading radio channels; linear filters; linear predictors; long-range prediction; mobile station; modeling properties; multipath mobile radio channel; nonlinear approach; prediction gain; quadratic Volterra filters; quadratic predictors; radio channel prediction; scatterers; wideband measurements; Delay; Fading; Gain measurement; Nonlinear filters; Optical scattering; Predictive models; Pulse shaping methods; Resource management; Stimulated emission; Wideband;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Vehicular Technology Conference, 1999. VTC 1999 - Fall. IEEE VTS 50th
  • Conference_Location
    Amsterdam
  • ISSN
    1090-3038
  • Print_ISBN
    0-7803-5435-4
  • Type

    conf

  • DOI
    10.1109/VETECF.1999.797075
  • Filename
    797075