How to implement Daubechies db4 from scratch in Matlab

Question

Sudhir Kumar 2022-4-20

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1700665-how-to-implement-daubechies-db4-from-scratch-in-matlab

回答： prabhat kumar sharma 2024-1-23

Is there any link to find Daubechies DB4 impliment from scratch in matlab?

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Sharmin Kibria 2023-8-18

Can you clarify the question a bit more? Are you looking for lifting steps or the filters for db4?

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

prabhat kumar sharma 2024-1-23

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1700665-how-to-implement-daubechies-db4-from-scratch-in-matlab#answer_1394966

在 MATLAB Online 中打开

Hi Sudhir,

You are trying to implement the Daubechies db4 wavelet from scratch in MATLAB can be a great learning experience, there are readily available implementations that can save your time and effort. Here are your options:

Signal Processing Toolbox: The Signal Processing Toolbox comes with built-in functions for various wavelets, including daubechies2 and daubechies4 that implement db2 and db4 wavelets, respectively. These functions provide efficient wavelet transform and inverse transform capabilities.
Wavelet Toolbox: If you need more advanced wavelet functionalities, consider the Wavelet Toolbox. It includes a broader range of wavelets and offers detailed tools for wavelet analysis, synthesis, and denoising.

Here is the basic minimal implementation idea of db4 wavelet decomposition:

function [cA, cD] = db4_wavelet_decompose(signal) 
    % Daubechies db4 wavelet decomposition 
    % signal: Input signal to decompose 
    % db4 decomposition low-pass and high-pass filter coefficients 
    Lo_D = [0.2303778133088964, 0.7148465705529154, 0.6308807679298587, ... 
           -0.0279837694168599, -0.1870348117190931, 0.0308413818355607, ... 
            0.0328830116668852, -0.0105974017850690]; 
    Hi_D = [-0.0105974017850690, -0.0328830116668852, 0.0308413818355607, ... 
             0.1870348117190931, -0.0279837694168599, -0.6308807679298587, ... 
             0.7148465705529154, -0.2303778133088964]; 
    
    % Convolve signal with low-pass and high-pass filters 
    cA = conv(signal, Lo_D, 'same'); 
    cD = conv(signal, Hi_D, 'same'); 
     
    % Downsample the convolved signals 
    cA = cA(1:2:end); 
    cD = cD(1:2:end); 
end 