• DocumentCode
    51374
  • Title

    Extraction of a Multi-Dimensional Polynomial Device Model for an Improved Distortion Contribution Analysis Technique

  • Author

    Aikio, Janne P. ; Rahkonen, Timo ; Pedro, Jose C.

  • Author_Institution
    Dept. of Electr. Eng., Univ. of Oulu, Oulu, Finland
  • Volume
    63
  • Issue
    1
  • fYear
    2015
  • fDate
    Jan. 2015
  • Firstpage
    155
  • Lastpage
    164
  • Abstract
    This paper presents a method to characterize a multi-dimensional polynomial current and charge source models needed for a distortion contribution analysis technique that can identify the dominant nonlinearities and different mixing mechanisms. A separate model characterization step with carefully chosen test signals is used to break the inherent input-output correlation that otherwise complicates the model fitting. The proposed Volterra-on-top-of-harmonic-balance technique and improved polynomial fitting are then verified through the analysis of a 100-W LDMOS RF Doherty amplifier, which showed that, in this circuit, the cubic nonlinearity of the gate capacitance is the main culprit of AM-PM distortion.
  • Keywords
    MIS devices; Volterra equations; harmonic distortion; multidimensional systems; polynomials; radiofrequency power amplifiers; AM-PM distortion; LDMOS RF Doherty amplifier; Volterra-on-top-of-harmonic-balance technique; cubic nonlinearity; distortion contribution analysis technique improvement; gate capacitance; input-output correlation; laterally diffused metal oxide semiconductor; mixing mechanisms; multidimensional polynomial current and charge source models; multidimensional polynomial device model extraction; polynomial fitting improvement; power 100 W; test signals; Analytical models; Correlation; Data models; Harmonic analysis; Integrated circuit modeling; Mathematical model; Polynomials; Distortion analysis; Doherty power amplifier (PA); Volterra analysis; polynomial device model;
  • fLanguage
    English
  • Journal_Title
    Microwave Theory and Techniques, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0018-9480
  • Type

    jour

  • DOI
    10.1109/TMTT.2014.2367530
  • Filename
    6963511