• DocumentCode
    1946263
  • Title

    Controllable parameters identification for high speed channel through signal-power integrity combined analysis

  • Author

    Choi, Myoung Joon ; Pandit, Vishram S. ; Ryu, Woong Hwan

  • Author_Institution
    Intel Corp., Folsom, CA
  • fYear
    2008
  • fDate
    27-30 May 2008
  • Firstpage
    658
  • Lastpage
    663
  • Abstract
    The optimization of high speed channel demands more challenging tasks such as estimating the noise from the interaction between signal nets and power nets, assessing the on-chip power delivery network (PDN) effectiveness, and including the power delivery (PD) to signal coupling noise into the channel budget. However, even just identifying what to optimize in high-speed channel is difficult task, and obtaining meaningful parameters including interaction between signal integrity and power integrity is more challenging. The proposed analysis method employs accurate and more effective ways to find controllable parameters to optimize the channel response for the best performance in the high speed channel considering both signal integrity (SI) and power integrity (PI) interactions by utilizing response decomposition in the time domain with worst case pattern consideration.
  • Keywords
    circuit noise; parameter estimation; switching circuits; time-domain analysis; channel response; high speed channel; on-chip power delivery network; parameter identification controllability; power integrity interactions; power nets; signal coupling noise; signal integrity; signal nets; signal-power integrity combined analysis; time domain decomposition; worst case pattern consideration; Cost function; Frequency; Parameter estimation; Performance analysis; Power grids; Power system modeling; Signal analysis; Signal processing; System-on-a-chip; Time domain analysis;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electronic Components and Technology Conference, 2008. ECTC 2008. 58th
  • Conference_Location
    Lake Buena Vista, FL
  • ISSN
    0569-5503
  • Print_ISBN
    978-1-4244-2230-2
  • Electronic_ISBN
    0569-5503
  • Type

    conf

  • DOI
    10.1109/ECTC.2008.4550042
  • Filename
    4550042