• DocumentCode
    478403
  • Title

    Stochastic Resonance in Saturation Nonlinearities

  • Author

    Zhang, L. ; Song, A.G.

  • Author_Institution
    Sch. of Instrum. Sci. & Eng., Southeast Univ., Nanjing
  • Volume
    5
  • fYear
    2008
  • fDate
    18-20 Oct. 2008
  • Firstpage
    607
  • Lastpage
    611
  • Abstract
    The majority of studies in saturation nonlinearities making use of periodic input signals and signal noise ratio (SNR) measures indicate the possibility of SNR gains due to stochastic resonance (SR). The paper points out that for such circumstances, the SNR measure discards some statistical information due to the saturation. When the shape of input signals is required at the output of the saturation nonlinearities, SNR is not an appropriate measure even for the case of narrowband or periodic signals. A statistical measure based on deflection ratio is proposed, and detectability at the output of the saturation nonlinearities is also considered simultaneously. Although SNR amplification has been shown to exist for periodic input signals, somewhat paradoxically, a gain larger than unity cannot be obtained as far as the deflection ratio or detectability is concerned. The underlying mechanism is revealed, and it is showed that the situations where the SNR gains occur are limited in their potential usefulness.
  • Keywords
    signal detection; statistical analysis; stochastic processes; deflection ratio; saturation nonlinearities; signal detection; signal noise ratio amplification; statistical measurement; stochastic resonance; Gain measurement; Gaussian noise; Instruments; Matched filters; Narrowband; Shape measurement; Signal detection; Signal to noise ratio; Stochastic resonance; Strontium; Saturating nonlinearity; Signal-to-noise ratio; Stochastic resonance; matched filter; signal detection;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Natural Computation, 2008. ICNC '08. Fourth International Conference on
  • Conference_Location
    Jinan
  • Print_ISBN
    978-0-7695-3304-9
  • Type

    conf

  • DOI
    10.1109/ICNC.2008.830
  • Filename
    4667507