• DocumentCode
    668931
  • Title

    Peak distortion analysis of nonlinear links

  • Author

    Beyene, Wendemagegnehu T.

  • Author_Institution
    Rambus Inc., Sunnyvale, CA, USA
  • fYear
    2013
  • fDate
    27-30 Oct. 2013
  • Firstpage
    169
  • Lastpage
    172
  • Abstract
    This paper introduces statistical analysis of high-speed nonlinear links by extending the peak distortion analysis to include the impact of nonlinear drivers. By taking advantage of the unique characteristics of interconnect networks, the multilinear theory is applied to calculate the responses of the nonlinear networks by analyzing a series of linear networks. First, the Volterra functional series is used to decompose the nonlinear link into multiple linear networks. Then, the peak distortion analysis of the high-speed nonlinear link is constructed from the linear combinations of sequences of peak distortion analyses of the decomposed linear networks. The analysis steps are derived analytically using an illustrative example. Finally, the accuracy of the method is verified for a high-speed memory link using time-domain simulation.
  • Keywords
    Volterra series; interconnections; linear network analysis; nonlinear distortion; nonlinear network analysis; statistical analysis; Volterra functional series; high-speed memory link; high-speed nonlinear links; interconnect networks; linear network analysis; multilinear theory; multiple linear networks; nonlinear drivers; nonlinear link decomposition; nonlinear network analysis; peak distortion analysis; statistical analysis; time-domain simulation; Analytical models; Integrated circuit interconnections; Integrated circuit modeling; Nonlinear distortion; Receivers; Time-domain analysis; Transmitters; Multilinear theory; Nonlinear I/O driver; Peak distortion analysis; Statistical analysis; Volterra series;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electrical Performance of Electronic Packaging and Systems (EPEPS), 2013 IEEE 22nd Conference on
  • Conference_Location
    San Jose, CA
  • Print_ISBN
    978-1-4799-0705-2
  • Type

    conf

  • DOI
    10.1109/EPEPS.2013.6703491
  • Filename
    6703491