• DocumentCode
    1722989
  • Title

    Formal error analysis and verification of a frequency domain equalizer

  • Author

    Souari, Anis ; Tahar, Sofiène ; Gawanmeh, Amjad

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Concordia Univ., Montreal, QC, Canada
  • fYear
    2012
  • Firstpage
    189
  • Lastpage
    192
  • Abstract
    In this paper we provide an approach for the formal verification of a frequency domain equalizer using higher-order logic based theorem proving. We perform a multi-level formal error analysis to verify an implementation of the equalizer based on the Fast LMS (Least Mean Square) algorithm. The formal error analysis is performed at the floating-point, fixed-point, and real numbers domains. The expressiveness of higher-order logic allows us to model the equalizer in all the three number domains and valuate the errors generated by approximating the floating-and fixed-point designs to the real domain of the frequency domain equalizer. This application shows the efficiency of formal methods in analyzing and verifying complex systems such as the frequency domain equalizer.
  • Keywords
    adaptive filters; equalisers; error analysis; formal verification; frequency-domain analysis; least squares approximations; theorem proving; adaptive filtering; fast LMS; fixed-point designs; floating-point designs; formal verification; frequency domain equalizer; higher-order logic based theorem proving; least mean square algorithm; multilevel formal error analysis; number domains; Algorithm design and analysis; Educational institutions; Equalizers; Error analysis; Frequency domain analysis; Least squares approximation; Signal processing algorithms;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    New Circuits and Systems Conference (NEWCAS), 2012 IEEE 10th International
  • Conference_Location
    Montreal, QC
  • Print_ISBN
    978-1-4673-0857-1
  • Electronic_ISBN
    978-1-4673-0858-8
  • Type

    conf

  • DOI
    10.1109/NEWCAS.2012.6328988
  • Filename
    6328988