Stochastic Resonance And Its Applications In Linear System With Asymmetric Dichotomous Noise

With the continual development of science and technology, requirements formeasuring & testing system and techniques are gradually improved. Requirements forhigh performance of measuring & testing system promote the progresses of its theories,methods and techniques. Linear system widely exists in variety of measuring & testingsystems. Problems about noise can't be avoided in measuring & testing system.Traditionally, we always try to reduce noise in order to improve system performance.Stochastic resonance (SR) presents us another thought, that is, noise, system andexternal force can also exist a cooperative effect on output performance. Based on thisidea, in this dissertation, stochastic resonance phenomena in linear system models areinvestigated, which are first-order and second-order linear systems with asymmetricdichotomous noise. And then, applying SR to improve performances of circuit modelswith noise is proposed. SR is verified by experiment in a linear system. It shows that SRcan improve output performance in a random linear circuit, and satisfactory results areachieved.The main research works and innovations in this dissertation are listed as follows:1. A simplified equivalence relation is proved in asymmetric dichotomous noise.The relation set up a connection between instantaneous power of noise and the intensity,asymmetry and instantaneous value of noise.2. The idea of parameter-induced stochastic resonance (PSR) is proposed infirst-order random linear system.3. Stochastic resonance is studied in a first-order linear system with asymmetricdichotomous noises, which are additive, multiplicative and bias signal modulated noiserespectively. The exact expressions of the first two moments and the signal-to-noiseratio (SNR) of the output are obtained. It is observed that the SNR dependsnon-monotonically on the parameters of noise, system and input signal.4. Stochastic resonance in a first-order linear system with multiplicative noisedriven by random amplitude periodical signal is investigated in case of asymmetricdichotomous noise. The conventional SR, the SR in a broad sense, the bona fide SR and PSR of SNR or output amplitude gain (OAG) are shown.5. The frequency characteristics and SRs of OAG in a second-order linear systemwith fluctuating intrinsic frequency or randomizing damping coefficient by asymmetricdichotomous noise are investigated. The results show that the OAG of system withnoise can be larger than without noise in a certain frequency band, and the asymmetryof noise affects the characteristic of the conventional SR.6. Applying SR to improve performances of three random linear circuit models withnoise is proposed, which are a RC series circuit disordered by time constant, a analogintegral circuit disturbed by digital signals and RLC low filter fluctuated byconductance.7. SR in a first-order linear system with asymmetric dichotomous noise is verifiedby experiment. The methods of applying PSR are presented in several first-orderrandom linear circuits.