Abstract:

Stochastic resonance can enhance the signal to noise ratio of the output signal greatly, which has attracted considerable attention in signal processing. Compared with the traditional noise induced stochastic resonance, parameter induced stochastic resonance improves the robustness. However, parameter induced stochastic resonance is confronted with the difficult problem of how to choose the best system parameter. Kurtosis and negentropy are introduced to measure the probability distribution function of the output signal of the nonlinear system. The relationship between the probability distribution characteristics and the best system parameter is found. A novel blind adaptive stochastic resonance method is proposed in this paper. This method is also applied in digital binary baseband binary signal processing. The proposed method utilizes kurtosis or negentropy to induce the iteration of system parameters, thus it tunes the nonlinear into stochastic resonance. The proposed method can solve the difficult problem of how to choose the best system parameter, which can enhance the robustness and agileness of stochastic resonance. The simulation platform is established with MATLAB to test the proposed method. The simulation results indicate the proposed method can converge to the best system parameter quickly, and it can improve the signal to noise ratio of the output signal greatly.

Key words: adaptive;digital baseband signal;probability distribution;stochastic resonance