Publication detail

Evolutionary Design of Transistor Level Digital Circuits using Discrete Simulation

MRÁZEK, V. VAŠÍČEK, Z.

Original Title

Evolutionary Design of Transistor Level Digital Circuits using Discrete Simulation

Type

conference paper

Language

English

Original Abstract

The objective of the paper is to introduce a new approach to the evolutionary design of digital circuits conducted directly at transistor level. In order to improve the time consuming evaluation of candidate solutions, a discrete event-driven simulator was introduced. The proposed simulator operates on multiple logic levels to achieve reasonable trade-off between performance and precision. A suitable level of abstraction reflecting the behavior of real MOSFET transistors is utilized to minimize the production of incorrectly working circuits. The proposed approach is evaluated in the evolution of basic logic circuits having more than 20 transistors. The goal of the evolutionary algorithm is to design a circuit having the minimal number of transistors and exhibiting the minimal delay. In addition to that, various parameter settings are investigated to increase the successrate of the evolutionary design.

Keywords

Evolutionary design, Transistor-level, Digital circuits, Cartesian genetic programming

Authors

MRÁZEK, V.; VAŠÍČEK, Z.

RIV year

2015

Released

15. 3. 2015

Publisher

Springer International Publishing

Location

Berlin

ISBN

978-3-319-16500-4

Book

Genetic Programming, 18th European Conference, EuroGP 2015

Edition

LCNS 9025

Pages from

66

Pages to

77

Pages count

12

URL

http://dx.doi.org/10.1007/978-3-319-16501-1_6

BibTex

@inproceedings{BUT119802,
  author="Vojtěch {Mrázek} and Zdeněk {Vašíček}",
  title="Evolutionary Design of Transistor Level Digital Circuits using Discrete Simulation",
  booktitle="Genetic Programming, 18th European Conference, EuroGP 2015",
  year="2015",
  series="LCNS 9025",
  pages="66--77",
  publisher="Springer International Publishing",
  address="Berlin",
  doi="10.1007/978-3-319-16501-1\{_}6",
  isbn="978-3-319-16500-4",
  url="http://dx.doi.org/10.1007/978-3-319-16501-1_6"
}