• DocumentCode
    610292
  • Title

    Comparison of passive enforcement techniques for DRAM package models

  • Author

    Elgamel, D. ; Barannyk, L. ; Elshabini, A. ; Barlow, F.

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Univ. of Idaho, Moscow, ID, USA
  • fYear
    2013
  • fDate
    12-12 April 2013
  • Firstpage
    25
  • Lastpage
    28
  • Abstract
    Passivity enforcement techniques are widely used in DRAM package model development. The Hamiltonian method is one of the common methods used for checking/enforcing passivity of these multiport networks. Enforcing the passivity for non-passive models is not trivial because it is built on several approximation steps starting from building the macromodel, which captures the frequency domain response, and ending with a suitable spice model for the DRAM circuits. Several approximation steps take place to achieve that result, and multiple constraints are used to preserve the original design characteristics. In this paper, we compared residue perturbation with and without error control in order to calculate how passivity enforcement affects the original model accuracy.
  • Keywords
    DRAM chips; DRAM circuit; DRAM package model development; Hamiltonian method; error control; frequency domain response; macromodel; multiport networks; nonpassive model; passive enforcement technique; passivity checking; passivity enforcement technique; residue perturbation; suitable spice model; Eigenvalues and eigenfunctions; Error correction; Mathematical model; Perturbation methods; Scattering parameters; Vectors; Absolute Error Control; Package Interconnects; Passivity; Residue Perturbation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Microelectronics and Electron Devices (WMED), 2013 IEEE Workshop on
  • Conference_Location
    Boise, ID
  • ISSN
    1947-3834
  • Print_ISBN
    978-1-4673-6034-0
  • Type

    conf

  • DOI
    10.1109/WMED.2013.6544510
  • Filename
    6544510