论文部分内容阅读
本文按照自成体系的离散小波变换正交分解框架,直接从离散域出发,得出离散小波变换域LMS(DWLMS)滤波器,变换的正交性与采样率无关,并可使用Mallat快速算法。该方法可减小自适应滤波器输入向量自相关阵的谱动态范围,提高了传统LMS算法的收敛速度和稳定性。
In this paper, according to the orthogonal decomposition framework of self-contained discrete wavelet transform, the discrete wavelet transform domain LMS (DWLMS) filter is derived directly from the discrete domain. The orthogonality of the transform is independent of the sampling rate, and the Mallat fast algorithm can be used. This method can reduce the spectral dynamic range of the auto-correlation matrix of the input vector of the adaptive filter and improve the convergence speed and stability of the traditional LMS algorithm.