• DocumentCode
    2462808
  • Title

    Researches on GRNN neural network in RF nonlinear systems modeling

  • Author

    Li-Na, Huang ; Jing-Chang, Nan

  • Author_Institution
    Sch. of Electrics & Inf. Eng., Liaoning Tech. Univ., Hu Lu Dao, China
  • fYear
    2011
  • fDate
    21-23 Oct. 2011
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    Simple introduces the structure of generalized regression neural network, describes the process that using generalized regression neural network (GRNN) modeling to RF nonlinear system, and through the simulation results demonstrate this new nonlinear modeling method of RF devices. The simulation results of the nonlinear RF devices by General regression neural network (GRNN) control the relative forecasting error within in 0.05 The best predict relative error by BP neural network is 0.17, so, compared with BP neural network, GRNN neural network effectively improved the accuracy and precision of the model. In the simulation process by general regression neural network (GRNN) method, to the nonlinear RF power amplifier device, the prediction curve before data normalized more close to the set output curve.
  • Keywords
    backpropagation; neural nets; nonlinear systems; regression analysis; BP neural network; GRNN neural network; RF nonlinear systems modeling; generalized regression neural network; Biological neural networks; Data models; Mathematical model; Predictive models; Radio frequency; Training; Vectors;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computational Problem-Solving (ICCP), 2011 International Conference on
  • Conference_Location
    Chengdu
  • Print_ISBN
    978-1-4577-0602-8
  • Electronic_ISBN
    978-1-4577-0601-1
  • Type

    conf

  • DOI
    10.1109/ICCPS.2011.6089937
  • Filename
    6089937