• DocumentCode
    2342779
  • Title

    A new macromodeling approach for nonlinear microwave circuits based on recurrent neural networks

  • Author

    Fang, Y.H. ; Yagoub, M.C.E. ; Wang, F. ; Zhang, Q.J.

  • Author_Institution
    Dept. of Electron., Carleton Univ., Ottawa, Ont., Canada
  • Volume
    2
  • fYear
    2000
  • fDate
    11-16 June 2000
  • Firstpage
    883
  • Abstract
    For the first time, recurrent neural networks (RNN) are trained to learn the dynamic responses of nonlinear microwave circuits. Once trained, the RNN macromodel provides fast prediction of the full analog behavior of the original circuit and can be used for high level simulation and optimization.
  • Keywords
    analogue circuits; circuit optimisation; circuit simulation; microwave circuits; nonlinear network analysis; recurrent neural nets; RNN macromodel; dynamic responses; full analog behavior; high level optimization; high level simulation; macromodeling approach; nonlinear microwave circuits; recurrent neural networks; Circuit simulation; Circuit topology; Design optimization; Differential equations; Equivalent circuits; Microwave circuits; Microwave devices; Neural networks; Nonlinear circuits; Recurrent neural networks;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Microwave Symposium Digest. 2000 IEEE MTT-S International
  • Conference_Location
    Boston, MA, USA
  • ISSN
    0149-645X
  • Print_ISBN
    0-7803-5687-X
  • Type

    conf

  • DOI
    10.1109/MWSYM.2000.863321
  • Filename
    863321