Parameter-induced stochastic resonance based on spectral entropy and its application to weak signal detection.

The parameter-induced stochastic resonance based on spectral entropy (PSRSE) method is introduced for the detection of a very weak signal in the presence of strong noise. The effect of stochastic resonance on the detection is optimized using parameters obtained in spectral entropy analysis. Upon processing employing the PSRSE method, the amplitude of the weak signal is enhanced and the noise power is reduced, so that the frequency of the signal can be estimated with greater precision through spectral analysis. While the improvement in the signal-to-noise ratio is similar to that obtained using the Duffing oscillator algorithm, the computational cost reduces from O(N(2)) to O(N). The PSRSE approach is applied to the frequency measurement of a weak signal made by a vortex flow meter. The results are compared with those obtained applying the Duffing oscillator algorithm.