论文部分内容阅读
传统基于小波分析的去噪方法和经验模态分解(EMD)去噪方法去噪后的信号信噪比较低。针对该问题,提出一种多小波包框架下区间迭代不变阈值的EMD去噪方法。对输入带噪信号进行预处理,将其变换为多维信号之后进行多小波包分解,设计针对软硬阈值函数的改进型阈值函数,并对得到的最后一层多小波包系数实现小波阈值处理,从而得到一维小波系数,对各本征模态函数分量(IMF)进行区间迭代不变阈值EMD去噪,并重构得到去噪后信号。仿真结果表明,与传统EMD小波阈值去噪方法相比,该方法信噪比提升近2.5dB,均方误差达到0.000 7,去噪效果较好。
The traditional denoising method based on wavelet analysis and the empirical mode decomposition (EMD) denoising method have low signal-to-noise ratio after denoising. To solve this problem, an iterative invariant threshold EMD denoising method under multi-wavelet packet framework is proposed. The input signal with noise is preprocessed, transformed into multi-dimensional signal, then multi-wavelet packet decomposition, the improved threshold function designed for hard and soft threshold function, and the last layer of multi-wavelet packet coefficients to achieve wavelet threshold processing, Thus one-dimensional wavelet coefficients are obtained. Interval invariant threshold EMD denoising is performed on each intrinsic mode function component (IMF), and the denoised signal is reconstructed. The simulation results show that compared with the traditional EMD wavelet threshold denoising method, the SNR of this method is improved by 2.5dB, the mean square error is 0.0007, and the denoising effect is better.