自适应小波包分解门限去噪新算法

提出了自适应小波包分解门限去噪的新方法。该方法自适应地对信号进行小波包分解,根据小波包子域的信噪比自适应选取去噪门限,并判定是否对该子域的信号进一步分解。与传统方法不同,新方法只需对不同尺度的部分概貌信号和细节信号根据该子域的信噪比大小进行分解,去噪后的信号按分解的逆过程进行重构。仿真结果表明,该方法相比于传统的小波去噪方法计算量有所降低,且去噪后的信号更接近真实的原始不含噪信号。

Adaptive Wavelet Package Decomposition Threshold De-noising Algorithm

This paper presents a new adaptive wavelet package decomposition threshold de-noising technique（AWPDD）. By this new method,signal is adaptively decomposed into wavelet package sub domains and according to the signal to noise ratio of each sub domain,we can determine the de-noising threshold and whether or not the sub domain should be decomposed further. Unlike the traditional methods,the new method only helps to decompose the partial contour and detail signal of some scales based on their signal to noise ratio. After de-noising,the signal is reconstructed by the inversely procedure of decomposition. Simulation results show the new method can lower the computing volume and the signal after de-noising is more close to the original no noise signal.