Code for multivariate denoising in wavelet toolbox

Question

Ashley Stephenson 2016-7-5

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/293702-code-for-multivariate-denoising-in-wavelet-toolbox

Hello, I am trying to process EMG signals using the multivariate denoising tool in the wavelet toolbox. I have been unable to export code from the GUI, as this doesn't appear to be an option for this tool, but I would like to duplicate the denoising process in my code in order to process many signals without having to use the GUI for each one. I have provided the relevant section of my code below, as well as a screen shot of the settings I would like to implement as shown in the toolbox.

for channelNum=1:1:8
            x = f.emgData(:,channelNum);
            % Set the de-noising method parameters. 
            level = 4;
            wname = 'db6';
            tptr  = 'minimaxi';
            sorh  = 's';
            % Set the PCA parameters
            npc_app = 1;
            npc_fin = 1;
            % Perform multivariate de-noising.
            [x_den, npc, nestco] = wmulden(x, level, wname, sym, npc_app, ... 
                                   npc_fin, tptr, sorh);         
              % Add filtered signals to filtered matrix
              filtered(:,channelNum) = x_den;

I am essentially trying to apply the settings shown below. When using the parameters above, I do not obtain the same results (filtered signal) as those generated by the GUI, and I'm not sure what I am doing differently. If anyone could help me with this I would appreciate it. Thanks.