• DocumentCode
    1837465
  • Title

    Periodicity enhancement of two-mode stochastic oscillators in a CNN type architecture

  • Author

    Mate, G. ; Horvath, E.A. ; Kaptalan, E. ; Tunyagi, A. ; Neda, Z. ; Roska, T.

  • Author_Institution
    Dept. of Phys., Babes-Bolyai Univ., Cluj-Napoca, Romania
  • fYear
    2010
  • fDate
    3-5 Feb. 2010
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    Two-mode stochastic oscillators coupled through a simple optimization dynamics exhibit interesting and novel collective behavior. Previous studies have shown that for a global coupling, apart of synchronization, a puzzling enhancement in the periodicity level of the global output can be achieved. Here we show that a locally coupled system which resembles a CNN-type architecture of the elementary units, is similarly able to exhibit synchronization for a given parameter range. Moreover, the resulting enhancement in the periodicity for a 2D lattice topology of the oscillatory units proves again to be significant. The system composed by non-perfect oscillators generates thus a perfectly periodic signal. An experimental realization of the system gives confidence in the numerical simulation results.
  • Keywords
    cellular neural nets; optimisation; oscillators; stochastic processes; synchronisation; CNN architecture; cellular neural network; optimization; periodicity enhancement; puzzling enhancement; synchronization; two-mode stochastic oscillators; Cellular neural networks; Frequency synchronization; Information technology; Lattices; Numerical simulation; Oscillators; Physics; Signal generators; Stochastic processes; Topology; CNN architecture; Monte-Carlo simulation; pulse-coupled oscillators; synchronization;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Cellular Nanoscale Networks and Their Applications (CNNA), 2010 12th International Workshop on
  • Conference_Location
    Berkeley, CA
  • Print_ISBN
    978-1-4244-6679-5
  • Type

    conf

  • DOI
    10.1109/CNNA.2010.5430275
  • Filename
    5430275