• DocumentCode
    2232555
  • Title

    Efficient derivation of exact s-expanded symbolic expressions for behavioral modeling of analog circuits

  • Author

    Shi, C. J Richard ; Tan, Xiangdong

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Iowa Univ., Iowa City, IA, USA
  • fYear
    1998
  • fDate
    11-14 May 1998
  • Firstpage
    463
  • Lastpage
    466
  • Abstract
    This paper presents an efficient method for deriving exact s-expanded symbolic expressions for behavioral modeling of analog circuits. The key idea is to introduce a graph-based representation called multi-root Determinant Decision Diagram (DDD), where each root represents the symbolic coefficient of a particular power of s. By exploiting sharing and sparsity, a huge number of symbolic product terms can be compactly represented and efficiently manipulated. Experimental results have demonstrated that this method can produce the exact s-expanded-symbolic transfer function for μA741 op amps in several CPU seconds on an UltraSparc-I workstation, and the use of generated symbolic expression for frequency-domain simulation achieved a speedup of about 50 over SPICE3F5
  • Keywords
    analogue integrated circuits; circuit simulation; directed graphs; frequency-domain analysis; integrated circuit modelling; symbol manipulation; μA741 op amps; UltraSparc-I workstation; analog circuits; behavioral modeling; frequency-domain simulation; graph-based representation; multi-root determinant decision diagram; s-expanded symbolic expressions; symbolic coefficient; transfer function; Analog circuits; Central Processing Unit; Circuit faults; Circuit simulation; Circuit topology; Cities and towns; Poles and zeros; Surges; Transfer functions; Workstations;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Custom Integrated Circuits Conference, 1998. Proceedings of the IEEE 1998
  • Conference_Location
    Santa Clara, CA
  • Print_ISBN
    0-7803-4292-5
  • Type

    conf

  • DOI
    10.1109/CICC.1998.695019
  • Filename
    695019