阵列调制随机共振在微弱信号特征提取方面的应用

Application of Array of Modulated Stochastic Resonance in Weak Signal Feature Extraction

论述了阵列调制随机共振方法在强噪声背景下多频微弱信号特征提取中的工作原理和实现步骤;采用预先设定系统参数的多个并联非耦合随机共振单元形成阵列,将被测强噪声背景下的多频微弱信号分别与不同频率的载波进行调制,生成多个差频均为0.01Hz的信号作为各对应随机共振单元的激励信号,采用龙格-库塔算法求取各单元输出并进行频谱分析,根据0.01Hz处的信噪比判断在微弱信号中是否存在载波频率与差频值之和大小的频率分量,最后综合各个随机共振单元的检测结果生成微弱信号的频率特征向量;仿真结果表明,阵列调制随机共振在微弱信号特征提取方面效果明显,具有很好的应用前景。

This paper discusses the principle and implementation steps of the method of array modulation stochastic resonance in feature extraction of multi--frequency weak signal under the background of strong noise. Using multiple parallel non--coupled resonant unit with preset system parameters to form an array, Making the measured multi--frequency weak signals respectively modulated with different fre- quency carrieres to generate multiple signals with the same difference frequency of 0.01Hz as the excitation signal of the corresponding sto- chastic resonance units, Getting the units outputs with Runge--Kutta algorithm and proceeding spectral analysis, then based on the signal to noise ratio （SNR） at 0.01Hz to determine whether there is the components whose frequency equals the value of carrier frequency plus the difference frequency in the weak signal. Finally, synthesizing the detection results of stochastic resonance units to generates the frequency feature vector of a weak signal. Simulation results show that the effect of weak signal feature extraction by modulation stochastic resonance array is obvious, and has a very bright application prospect in the future.