Analysis of Bistable System Noise and Study for Weak Signal Detection

It is investigated that the noise energy in the power spectrum is concentrated in the low frequency region when a bistable system responds to the noise. In terms of the adiabatic elimination theory, the viewpoint is deduced that the main stochastic resonance (SR) peak is only produced in the low frequency region. With the new method of frequency compression transformation stochastic resonance (FCTSR) proposed here, the application of the adiabatic elimination principle for detecting a weak signal embedded in noise is analyzed under the condition of large parameters. The numerical simulation shows that, so long as the input signal-to-noise ratio of the bistable system is larger than -30 dB and the real practical frequency is not smaller than 50 times of the value of the weak periodic signal frequency, the SR spike at the signal frequency can be obtained from the response spectrum of the bistable system, and hence the weak signal submerged in strong noise is able to be extracted.