Engineering signal processing based on adaptive step-changed stochastic resonance

Weak signal detection which is under the condition of adiabatic elimination in large parameters can be solved by step-changed stochastic resonance (SCSR) presented by our group. Adaptive SCSR based on approximate entropy (ApEn) is also proposed in this paper, and it can get the best result of SCSR adaptively. Our analysis shows that the ApEn value of periodic signal is related to its frequency and signal-to-noise ratio (SNR), but not to the change of its amplitude and phase. So a periodic signal with definite SNR whose frequency is to be detected can be made under the same sampling condition as the raw data, and its ApEn is calculated as a standard reference. By adjusting the structural parameters and calculation step automatically, a series output of the bistable system can be got, and an ApEn distance matrix can be constructed. After getting the minimum value of the matrix, the best parameters of the non-linear system and calculation step can be obtained. Two examples of detecting weak signal mixed with heavy noise are presented in the end to illustrate that SCSR and its adaptive solution are effective for signal processing.