In nonlinear system, it is of great significance in engineering practice to achieve stochastic resonance, which is a nonlinear method used for weak signal detection. In view of the highfrequency weak signal with unknown background noise in industry field, the parameterinduced adaptive stochastic resonance based on parameter compensation is proposed. The system parameters are optimized by knowledgebased particle swarm optimization in parallel, which takes the output signalnoiseratio of system as the fitness function and the property that stochastic resonance system produces the best resonance effect just when the intensity of noise approximately equals the potential barrier as the knowledge. Finally the adaptive stochastic resonance system based on DSP is designed, which can process the signals in real time. And the system can display the detection results in the touch screens in real time, realizing the weak signal detection successfully and completely.