• DocumentCode
    594062
  • Title

    A mixed-domain behavioral model´s extraction for digital I/O buffers

  • Author

    Dghais, Wael ; Cunha, Telmo R. ; Pedro, Jose C.

  • Author_Institution
    Inst. de Telecomun., Univ. de Aveiro, Aveiro, Portugal
  • fYear
    2012
  • fDate
    21-24 Oct. 2012
  • Firstpage
    224
  • Lastpage
    227
  • Abstract
    The paper presents a novel extraction procedure based on the frequency domain formulation of the current-charge (I-Q) behavioral model for digital I/O buffers/drivers output admittance followed by a time domain extraction of the predriver´s nonlinear dynamic functions. The large signal model´s functions of the drivers´ output admittance are derived from the bias-dependent scattering, or S-parameters, measurements. This easy and fast extraction method allows the accurate generation of the data-based model from automated and straightforward measurements. The extracted model´s functions are implemented as lookup tables (LUTs) and the behavioral model is validated in typical SI scenario.
  • Keywords
    S-parameters; buffer circuits; driver circuits; table lookup; I-Q behavioral model; LUT; S-parameter measurement; bias-dependent scattering; current-charge behavioral model; data-based model; digital I-O buffer-driver output admittance; frequency domain formulation; large signal model function; lookup table; mixed-domain behavioral model extraction; predriver nonlinear dynamic function; time domain extraction; Admittance; Computational modeling; Data models; Frequency domain analysis; Integrated circuit modeling; Predictive models; behavioral models extraction; digital I/O buffer; large signal model; signal integrity;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electrical Performance of Electronic Packaging and Systems (EPEPS), 2012 IEEE 21st Conference on
  • Conference_Location
    Tempe, AZ
  • Print_ISBN
    978-1-4673-2539-4
  • Electronic_ISBN
    978-1-4673-2537-0
  • Type

    conf

  • DOI
    10.1109/EPEPS.2012.6457882
  • Filename
    6457882