Aiming at the limitation of stochastic resonance (SR) system for dealing with complex signals and the blindness of parameter selection, a stochastic resonance method matching with quantum particle swarm adaptive parameter is presented based on frequency information exchange (FIE). Firstly, FIE is used to exchange the frequency amplitude information of the high-frequency characteristic signal to the corresponding reference low frequency. Then, according to the reference frequency, the system parameters of characteristic SR are optimized by using the quantum particle swarm optimization (QPSO) algorithm. Finally, the vibration signal is processed by stochastic resonance method. The measured signal analysis of rolling bearing shows that this method can eliminate the limitation of stochastic resonance on frequency band and avoid the blindness of parameter selection, which makes the stochastic resonance more suitable for the detection of high-frequency band fault signal in strong noise background.