Abstract: As traditional stochastic resonance methodhas low accuracy problem for weak signal detection,a new adaptive stochastic resonanceweak signal detection method based on global artificial fish swarm algorithm (GAFSA) is proposed, in which the global artificial fish swarm algorithm is combined with stochastic resonance. The output signal-to-noise ratio gain of stochastic resonance system is used as fitness function of the global artificial fish swarm algorithm, and the weak signal detection problem is transformed into multi-parameter parallel optimization problem, which formsan adaptive system to detect weak feature signal. The simulation experiments and the comparisons of optimization results are carried out respectively in the Langevin and Duffing systems. Simulation results show thatthe proposed method is efficient and feasible. Compared to Langevin system, Duffing adaptive stochastic resonance system has higher detection accuracy and performance of the weak signal detection. Duffing optimal system is applied to the multi-frequency large signal detection, which widens the application range of stochastic resonance.

Key words: self-adaption, stochastic resonance, Duffing oscillator, global artificial fish swarm algorithm, weak signal detection