Stochastic Resonance Based On Non-uniform Periodic Sampling

Real-time monitoring of the critical equipment's running state is related to the development of the national economy and national construction. Along with the development of science and technology and the improvement of the engineering requirement, the real-time monitoring puts forward higher request for the method of detection and signal processing in order to detect the weak fault characteristic signal in strong noise environment specially.Scale-transformation stochastic resonance technology can use the collaborative relationship of bistable system, noise and signal to strengthen the weak characteristic signal and then to detect, extract and identify it. But it need to make the system parameters, noise intensity, signal parameters and sampling frequency of variable scale achieve a tuned relationship. The process of adjusting parameters is complex relatively, especially it requires the sample frequency 50 times higher than the characteristic signal frequency, so it limits the application of the scale-transformation stochastic resonance in the engineering practice to some extent.The non-uniform periodic sampling technology can sample the high frequency signal using lower sampling frequency without producing spectrum aliasing, and can also eliminate the aliasing noise generated by completely non-uniform sampling. This paper presents a new method of nonlinear signal processing: Stochastic Resonance Based on Non-uniform Periodic Sampling, by combining periodic non-uniform sampling and stochastic resonance. This method neither requires excessive parameters adjustment, nor requires a sample frequency 50 times higher than the signal frequency, it just need to fix one group stochastic resonance system parameters, use a group of lower and different sampling frequencies to sample, through the bistable system, and then use the Fourier transform of the non-uniform sampling to superpose in frequency spectrum. It can detect the weak characteristic signal clearly. This approach makes stochastic resonance technology more widely applied in engineering practice. The computer simulation and engineering test demonstrate its efficiency.