Analyzing distortion contributions in a complex device model

• Timo Rahkonen ^[1] ; Janne P. Aikio ^[1]
  1. [1] University of Oulu

     University of Oulu

     Oulu, Finlandia

     
• Localización: Compel: International journal for computation and mathematics in electrical and electronic engineering, ISSN 0332-1649, Vol. 33, Nº 4 (Special Issue: SCEE 2012), 2014, págs. 1264-1271
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Purpose – The purpose of this paper is to propose a method to reduce the non-linear distortion of a transistor to its input and output ports to aid distortion contribution analysis (DCA). This is especially needed when the internal structure of a device model is complex.

    Design/methodology/approach – The non-linear distortion generated by all non-linear sources inside a device model are reduced to transistor i/o ports by LMSE fitting techniques. Simulations of an LDMOS power transistor are used to compare the reduced distortion results with the actual non-linear sources.

    Findings – It is shown, that device models where the current sources are split by intermediate nodes cause superficial results, when distortion contributions are calculated as a superposition of contributions from individual non-linear sources. The proposed iterative fitting technique works.

    Research limitations/implications – Some non-quasistatic effects and the transfer functions from external terminals to internal controlling nodes are not covered.

    Practical implications – The analysis is a step toward a generic non-linear distortion contribution simulation tool that would aid the designers to develop more linear analog circuits.

    Originality/value – The concept of DCA itself is fairly new. This paper makes a step to represent the distortion sources in a canonical way.