• DocumentCode
    684283
  • Title

    A new representation method based on mapping functions for analog circuit automatic design

  • Author

    Yong Zheng ; Jingsong He

  • Author_Institution
    Dept. of Electron. Sci. & Technol., Univ. of Sci. & Technol. of China, Hefei, China
  • fYear
    2013
  • fDate
    19-21 Oct. 2013
  • Firstpage
    171
  • Lastpage
    176
  • Abstract
    Circuit evolutionary design is a way to use evolutionary algorithms to design circuits automatically. Designing a circuit through this method, we must code the circuit first, representing a circuit by a chromosome. This paper proposes an analog circuit coding method based on mapping functions. We use two real numbers between 0 and 1 to represent a component. Mapped by the predefined functions, the two numbers can represent the component type and its connection nodes, they can also represent the component parameter if necessary. The adjacent two components in the chromosome share one-digit code. The coding method has the following features: the chromosome length is nearly equal to the number of components in the circuit, no matter how many components the circuit has; the syntax of the coding is closed, so it won´t produce any illegal circuit; the structure of the circuit represented by this coding method is very rich. At last, we conduct two circuit design experiments using the function coding method proposed in this paper, designing a circuit only contain two-terminal components and a circuit containing three-terminal components. We select a typical circuit-a passive filter only containing L and C and an amplifier containing T as our target circuits. The experiment results show that the proposed coding method is effective for designing circuits.
  • Keywords
    circuit optimisation; encoding; evolutionary computation; network synthesis; amplifier; analog circuit automatic design; analog circuit coding method; chromosome length; circuit evolutionary design; component parameter; connection nodes; evolutionary algorithms; function coding method; mapping functions; one-digit code; passive filter; representation method; three-terminal components; two-terminal components; Biological cells; Encoding; Transistors;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Advanced Computational Intelligence (ICACI), 2013 Sixth International Conference on
  • Conference_Location
    Hangzhou
  • Print_ISBN
    978-1-4673-6341-9
  • Type

    conf

  • DOI
    10.1109/ICACI.2013.6748496
  • Filename
    6748496