基于随机共振电路模拟的微弱周期信号检测

采用电路模拟非线性Duffing振子，利用其随机共振机制来检测微弱周期信号。针对随机共振只适用于极低频输入信号的限制，引入一种适当的变量变换可以将高频信号转化成符合随机共振理论要求的低频信号进行处理，增强了该方法在工程应用中的可行性。采用电路模拟方法检测微弱周期信号，不需要象随机共振数值仿真所要求的那样对信号过采样，在满足采样定理的条件下，可以取较小的采样频率，降低了对硬件的要求。实验表明，该方法能有效地从强背景噪声中检测出微弱周期信号，在机械系统故障早期检测、化学谱信号提取、多传感器测量等领域有实际应用价值。

Detecting weak signal based on analog simulating of stochastic resonance

A scheme of circuit simulation for Duffing oscillator is proposed. The mechanism of stochastic resonance (SR) with associated oscillator is used to detect weak periodic signal. Because only those signals with extremely low frequency can be processed by SR theory, a scheme based on valuable substitution is developed to transform the input signal into its substituent with lower frequency to meet this requirement. Accordingly, restriction by conventional SR theory is broken through, allowable frequency of the input signal becomes much higher, and the input signal is no longer to be over-sampled during the process of digitization. As the result, feasibility of the stochastic resonator is improved for practical situations. Experiment results show that the scheme is capable of detecting weak sinusoidal signal from the intense noise background, which is of practical importance in incipient fault detection of mechanical systems, weak chemical spectrum signal detection and multi-sensor measurement.