基于奇异值分解的随机共振特征提取研究

Features extraction based on singular value decomposition and stochastic resonance

针对强背景噪声下信噪比极低的微弱特征信号的识别问题,提出了基于奇异值分解的随机共振特征提取方法.该方法首先利用奇异值分解对实际采样信号进行预处理和重构,然后寻找到特征信号分量与噪声强度相匹配的分量信号.此分量信号再经过非线性双稳系统的随机共振处理,可实现从强噪声背景中检测极微弱的特征信号.

In order to detect the weak characteristic signal submerged in heavy noise with extremely low signal-to-noise ratio, a method based on singular value decomposition （SVD） and stochastic resonance is proposed. The sampling signal is first preprocessed and reconstructed by means of SVD, and then we search for a component signal. In the component signal, the components of the characteristic signal match noise strength. Then the component signal is processed with the non-linear bistable system to obtain stochastic resonance response, thus the goal of detecting the weak characteristic signal submerged in a heavy background noise is realized.