• [email protected]
  • +971 507 888 742
Submit Manuscript
SciAlert
  • Home
  • Journals
  • Information
    • For Authors
    • For Referees
    • For Librarian
    • For Societies
  • Contact
  1. Information Technology Journal
  2. Vol 12 (4), 2013
  3. 841-846
  • Issues
    Online First Current Issue All Issues
  • Information About
    Aims and Scope Editorial Board Guide to Authors Article Processing Charges
    Submit a Manuscript

Information Technology Journal

Year: 2013 | Volume: 12 | Issue: 4 | Page No.: 841-846
DOI: 10.3923/itj.2013.841.846
crossmark

Facebook Twitter Reddit Linkedin E-mail
Research Article

Research on Stochastic Resonance of Large Parameters Periodic Signal

Han Lingyun
Institute of Millimeter Wave and Optical Near Sensing Technology, School of Electronic Engineering and Optoelectronic Technology, Nanjing University of Science and Technology, Nanjing 210094, China

Li Yuehua
Institute of Millimeter Wave and Optical Near Sensing Technology, School of Electronic Engineering and Optoelectronic Technology, Nanjing University of Science and Technology, Nanjing 210094, China

ABSTRACT


For the question of frequency scale of stochastic resonance of bi-stable system, this study proposed a novel prove based on parameter normalized transformation and clear explanations viewed from physical meaning is presented. Experiments of large parameters periodic signal of stochastic resonance based on parameter normalized transformation principle clarified that this principle is applied to periodic signal of any frequency. The results of this study have positive practical value to large parameter weak periodic signal recovery and enhance transmission.
PDF Abstract XML References Citation

Keywords


  • Bi-stable system
  • weak periodic signal recovery
  • large parameter stochastic resonance

Article History

Received: March 26, 2012;   Accepted: December 15, 2012;   Published: January 22, 2013

How to cite this article

Han Lingyun and Li Yuehua, 2013. Research on Stochastic Resonance of Large Parameters Periodic Signal. Information Technology Journal, 12: 841-846.

DOI: 10.3923/itj.2013.841.846

URL: https://scialert.net/abstract/?doi=itj.2013.841.846

INTRODUCTION

Initially, stochastic resonance is the concept put forward in the seminal papers by Benzi et al. (1981) to address the problems of periodic occurrence of earth’s ice ages, abrupt glacial changes, signal to noise ratio in crayfish mechanoreceptors or noise in neurons (Benzi et al., 1981, 1983). It describes a bi-stable system with nonlinear, under the action of a small periodic modulation signal and inputting noise simultaneously, when the noise increases to a certain intensity, the system output SNR will greatly boost and exist a best input noise intensity which can make the system produces a highest output SNR.

Godivier and Chapeau-Blondeau (1997) and Apostolico (1998) proved the existence of the stochastic resonance phenomenon in the Schmitt trigger circuit systems with two steady state output characteristics. Dykman and McClintock (1998) proposed that SNR gain of nonlinear system can be greater than 1. Petracchi (2000) compared the traditional signal processing recovery method and the stochastic resonance method, supported the Dykman and McClintock (1998) standpoint. Another question is about frequency, Kaufman (1996) put forward high-frequency stochastic resonance. Chen and Varshney (2008), Martinez et al. (2011) and Duan (2011) presented instances of recovery of high frequency periodic signal. There is no clear exposition from available literature about the frequency scale applied in stochastic resonance of the bi-stable system in theory and physics. This study confirmed that weak periodic signal recovery based on stochastic resonance of bi-stable system doesn’t limited to frequency and that for large parameter stochastic resonance the hybrid input signal must also have greater amplitude.

THEORY AND MODEL OF STOCHASTIC RESONANCE OF BI-STABLE SYSTEM

Stochastic resonance is a kind of nonlinear phenomena, in which noise interference signal applied to a class of nonlinear system, signal and noise in nonlinear system on the synergy effect, noise energy will be transferred to signal energy, produce similar to mechanics resonance output, so as to improve the signal-to-noise ratio to achieve the purpose of weak signal recognition.

Three basic elements to realize stochastic resonance are nonlinear system with potential barrier, inner or outer noise source and input signal. Bi-stable system is a typical nonlinear system with wide range applications in physical, chemistry and other natural science.

Langevin equation is the most typical one to describe stochastic resonance of nonlinear bi-stable system (Gammaitoni and Marchesoni, 1989):

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
(1)

where a, b are the structural parameters of potential well and are real number and S(t) = Acos (2πf0t) is measured signal, A is the amplitude and f0 is frequency and Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal is Gauss white noise, D is noise strength, g (t) is Gauss white noise with mean value 0 and variance 1. Noise Γ (t) meets the condition of statistical average <Γ (t)> = 0, <Γ (t')> = 2Dδ (t-t '). Bi-stable system stochastic resonance model is shown in Fig. 1, where, x (t) is output signal, nonlinear bi-stable system is expressed in potential function U (x) = -a/2x2+b/4x4. Potential function curve is shown in Fig. 2 (the simulation assumption is a = b = 1).

• A = 0, Γ (t) = 0, given initial value x0>0 or x0<0, when t74, X (t) will be infinite approximation to potential well Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal or Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal , also, two potential well be separated by a potential barrier highly ΔV = a2/4b. The barrier height described the potential difference need to overcome when transition from one potential well to another. When x0 = 0, system output is x (t) = 0
• A = 0, Γ (t)≠0, x (t) is in transition switching in two potential well with Kramer transition rate rk, rk depends on the noise distribution and intensity D which calculation formula is:

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
(2)

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
Fig. 1: The model of bi-stable stochastic resonance system

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
Fig. 2: The curve of potential function of bi-stable system

• A≠0, Γ (t)≠0, when A,Ac (Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal is the bi-stable system static trigger threshold value), x (t) is in potential well Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal or Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal with input frequency f0; when A.Ac, x (t) will overcome potential barrier and switch output between potential well with input frequency f0. But according to adiabatic approximation theory, in the small parameter signal conditions, in other words, when 0<f0≤1,0<D≤1, though A,Ac, even A≤Ac, x (t) will be switch output between potential well with input frequency f0, part of noise energy trans into signal energy, coursing two potential well position difference ΔV far overweigh the input signal altitude, effectively enlarge the system output signal altitude, so as to increase the signal-to-noise ratio of the output signal

PARAMETER NORMALIZED TRANSFORMATION OF BI-STABLE SYSTEM

In the traditional idea, stochastic resonance only occurs when signal parameter is small, namely amplitude and frequency of signal also far less than 1. But in engineering practice it is often needed to deal with signal with frequency far outweigh 1, so how to detect high-frequency weak periodic signal making use of stochastic resonance has important meaning.

In the study of stochastic resonance, usually take a = b = 1. For example, take a = 1, b = 1, A = 0.3, f0 = 0.01 Hz sampling frequency fs = 5 Hz, when D = 0.31 bi-stable system achieve stochastic resonance state, as shown in Fig. 3. Runge-Kutta algorithm is used for numerical calculation experiment, 4000 sampling point, the system input and output signal time domain and spectrum diagram is shown in Fig. 3, where spectrum diagram amplitude normalized processing.

From output signal of the Fig. 3, we can see a good stochastic resonance phenomenon and that the weak periodic signal has been strengthened greatly, namely that the noise energy transferred to be signal energy. This results show that by bi-stable system stochastic resonance signal-to-noise ratio of small parameter weak periodic signal can be effectively enhanced. In order to use bi-stable system to realize stochastic resonance of weak periodic signal of big parameters, firstly introduced parameters normalized transformation of Eq. 1.

When a≠1, b≠1 of Eq. 1, we can do the transformation as follows:

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
(3)

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
Fig. 3(a-d): Stochastic resonance wave and frequency spectrum of bi-stable system when f0 = 0.01 Hz, a = 1, b = 1, (a) Signal wave of input, (b) Frequency spectrum of input, (c) Signal wave of output and (d) Frequency spectrum of output

Substituted to Eq. 1, we have:

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
(4)

Equation 4 also can be:

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
(5)

As seen in Eq. 1, Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal meets the condition of statistical average <Γ (t)> = 0, <Γ (t) Γ (t')> = 2Dδ (t-t'). Signal Γ (t) is Gaussian white noise; in the power spectrum density distribution diagram power spectrum curve is constant and don’t change with frequency, so, Γ (τ/a) is also Gaussian white noise and meets the condition of statistical average <Γ (τ/a)> = 0, <Γ (τ/a) Γ (τ'/a)> = 2Dδ (τ/a-τ'/a), namely:

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
(6)

So Eq. 5 can be:

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
(7)

Equation 7 is equivalent to Eq. 1. Theoretically, from Eq. 7 we can see that, for big parameter periodical signal, if parameter a is big enough, high frequency periodic signal can be transformed to low frequency periodic signal and meet the condition of small parameters and suitable parameter b can be choose to different amplitude of input high frequency periodic signal.

EXPERIMENTS OF LARGE PARAMETERS PERIODIC SIGNAL STOCHASTIC RESONANCE

Without loss of generality, assume that in large parameter system high frequency of periodic signal is f0 = 1000 Hz, according to the parameter normalized transformation principle, other parameters of bi-stable system can be derived as a = b = 105, fS = 500 kHz, if the periodic signal amplitude and noise strength remain unchanged, also A = 0.3, D = 0.31, we can see that stochastic resonance happened in the bi-stable system, as in Fig. 4.

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
Fig. 4(a-d): Stochastic resonance wave and frequency spectrum of bi-stable system when f0 = 1 kHz, a = 105, b = 105, (a) Signal wave of input, (b) Frequency spectrum of input, (c) Signal wave of output and (d) Frequency spectrum of output

From Fig. 4b, we can see that the output signal is transiting to and fro between two steady value of +1 and -1, its wave is similar with Fig. 3c, this is characteristics of stochastic resonance happened in bi-stable system. So, under the large parameter station (f0 = 1000 Hz, a = b = 105, fS = 500 kHz), stochastic resonance happened in bi-stable system.

According to the parameter normalized transformation principle, if f0 = 1000 Hz is determined, the bi-stable system parameter ‘a’ is also determined a = 105 but this is not suitable for parameter ‘b’, ‘b’ can be not equal to ‘a’ and can be any number greater than zero. Also for example as f0 = 1000 Hz, assume that another bi-stable system with parameter a = 105, b = 100, amplitude of input periodic signal A = 0.3 and noise strength D = 0.31 remain unchanged, the sum of which is the system input signal. According to the parameter normalized transformation principle, amplitude of the sum signal will be amplified Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal times, substituting Eq. 1; the result can be shown in Fig. 5. In this instance two steady state value of bi-stable system is Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal.

From Fig. 5c we can see that the output signal is transiting to and fro between two steady value of +10 and -10, stochastic resonance happened in bi-stable system.

From Fig. 4 and 5 we can see that, stochastic resonance happened in bi-stable system when parameter ‘b’ is different, output signal wave is same in addition to different steady state value and the steady state value of the output signal is the same with theoretical value. So, stochastic resonance of bi-stable system of large parameters can be transformed to one system of small parameters according to the parameter normalized transformation principle and the periodic signal frequency can take arbitrary value.

The above analysis and experiments show that for high frequency periodical signal, bi-stable system parameter and input signal amplitude can be adjusted to suitable for stochastic resonance condition and realize weak signal recovery. To realize the high frequency of periodic signal stochastic resonance, the key is that the parameter ‘a’ of bi-stable system is large enough to make normalized periodic signal frequency is far less than 1 to comply the small parameter condition.

Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal
Fig. 5(a-d): Stochastic resonance wave and frequency spectrum of bi-stable system when f0 = 1 kHz, a = 105, b = 1000, (a) Signal wave of input, (b) Frequency spectrum of input, (c) Signal wave of output and (d) Frequency spectrum of output

This can also be explained with Kramer’s transition rate (escape rate) of bi-stable system. For bi-stable system, if only drive by noise Γ (t) without periodical signal, x (t) is in transition switching in two potential well with Kramer transition rate Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal, ΔU = a2/4b is the barrier height of bi-stable system. When noise strength D tends to infinity, Image for - Research on Stochastic Resonance of Large Parameters Periodic Signal is its limit value, showing that transition rate between two potential well of bi-stable system cannot be more than this value. When frequency of periodical excitation signal matches half of the transition rate f0 = rk/2, stochastic resonance phenomenon will occur. So, frequency able to generate stochastic resonance is limited in 0<f0.<rklim/2 and Kramer’s transition rate is the main factors to restrict the phenomenon of stochastic resonance of the high frequency periodic signal. It also shows that for high frequency periodic signal, the bi-stable system parameter ‘a’ is a key factor which is confirmed by above analysis and experiments.

CONCLUSION


According to nonlinear stochastic resonance principle of bi-stable system used in high frequency weak periodic signal recovery, parameters normalized transformation principle is derived. And experiments about stochastic resonance of large parameters periodic signal are preceded which show that weak periodic signal recovery based on stochastic resonance doesn’t limited to frequency and that for large parameter stochastic resonance the hybrid input signal must also have greater amplitude. The results of this paper have positive practical value to large parameter weak periodic signal recovery and enhance transmission.

ACKNOWLEDGMENTS


The authors would like to thank the anonymous reviewer and editors for their helpful comments and suggestions. This study is supported by National Natural Science Foundation of China under Grants 60901008 and 61001010, National Ministry Foundation of China under Grants 9140A05070910BQ02 and 51305050102.

REFERENCES


  1. Benzi, R., A. Sutera and A. Vulpiani, 1981. The mechanism of stochastic resonance. J. Phys. A: Math. Gen., 14: 453-457.
    CrossRefDirect Link

  2. Benzi, R., G. Parisi, A. Sutera and A. Vulpiani, 1983. Theory of stochastic resonance in climatic change. SIAM J. Applied Math., 43: 565-578.
    Direct Link

  3. Godivier, X. and F. Chapeau-Blondeau, 1997. Noise-assisted signal transmission in a nonlinear electronic comparator: Experiment and theory. Signal Proc., 56: 293-303.
    CrossRefDirect Link

  4. Apostolico, F., L. Gammaitoni and S. Santucci, 1998. Color effects in a near-threshold schmitt trigger. Phys. Rev. E, 58: 7079-7084.
    Direct Link

  5. Dykman, M.I. and P.V.E. McClintock, 1998. What can stochastic resonance do?. Nature, 391: 344-344.
    Direct Link

  6. Petracchi, D., 2000. What is the role of stochastic resonance?. Chaos Solitons Fractals, 11: 1827-1834.
    CrossRefDirect Link

  7. Kaufman, I.K., D.G. Luchinsky, P.V.E. McClintock, S.M. Soskin and N.D. Stein, 1996. High-frequency stochastic resonance in SQUIDs. Phys. Lett. A, 220: 219-223.
    CrossRefDirect Link

  8. Chen, H., P.K. Varshney and J.H. Michels, 2008. Improving sequential detection performance via stochastic resonance. IEEE Signal Process. Lett., 15: 685-688.
    CrossRef

  9. Martinez, E., G. Finocchio and M. Carpentieri, 2011. Stochastic resonance of a domain wall in s stripe with two pinning sites. Applied Phys. Lett., 98: 072507-072507-3.
    CrossRef

  10. Duan, M., 2011. A new albedo model constructed for stochastic resonance simulation on 100ka climate cycle. Proceedings of the International Conference on Information Science and Technology, March 26-28, 2011, Nanjing, pp: 132-135.
    CrossRef

  11. Gammaitoni, L. and F. Marchesoni, 1989. Stochastic resonance in Bi-stable systems. Phys. Rev. Lett., 62: 349-352.
    Direct Link

Related Articles

Leave a Comment


Your email address will not be published. Required fields are marked *

Useful Links

  • Journals
  • For Authors
  • For Referees
  • For Librarian
  • For Socities

Contact Us

Office Number 1128,
Tamani Arts Building,
Business Bay,
Deira, Dubai, UAE

Phone: +971 507 888 742
Email: [email protected]

About Science Alert

Science Alert is a technology platform and service provider for scholarly publishers, helping them to publish and distribute their content online. We provide a range of services, including hosting, design, and digital marketing, as well as analytics and other tools to help publishers understand their audience and optimize their content. Science Alert works with a wide variety of publishers, including academic societies, universities, and commercial publishers.

Follow Us
© Copyright Science Alert. All Rights Reserved