Trouble with for loop to determine the discrete wavelet transform (DWT)

Hi, I need assistance plotting the approximation (A) and detail (D) coefficients separately for all sheets in an Excel file over a given time frame. Currently, the existing code generates the transform only for the final sheet (Sheet 8). I am uncertain how to modify the code to plot A1 and D1 separately for all sheets. I would be grateful for any guidance you may provide. Thank you for your time and help.

clc
close all
clear all

filename = 's.xlsx';

% Get the sheet names
[~, sheetNames] = xlsfinfo(filename);

for i = 1:numel(sheetNames)
    % Read the data from the sheet
    data = xlsread(filename, i);
    % Extract the time and signal columns
    t(:,i) = data(:, 1);
    sig (:,i) = data(:, 2);
    N = length(sig);
    dt = t(3) - t(2); % sampling time
    fs = 1/dt; % freq
    signal = sig (:,i);
    wname = 'bior6.8';
    % Discrete Wavelet, DWT
    [CA1,CD1] = dwt(signal,wname);
    A1 = idwt(CA1,[],wname,N);
    D1 = idwt([],CD1,wname,N);

    t (:,i) = linspace(0,N,N)*(1/fs);
    subplot(1,2,1);
    plot (t,A1,'k','LineWidth',1.5);
    title(sprintf('Approximation for sheet%d', i));
    set(gca,'fontname','Times New Roman','FontSize',10)
    xlabel('Time (secs)')
    ylabel('Amplitude')
    grid on
    subplot(1,2,2);
    plot (t,D1,'k','LineWidth',1.5);
    title(sprintf('Detail for sheet%d', i));
    set(gca,'fontname','Times New Roman','FontSize',10)
    xlabel('Time (secs)')
    ylabel('Amplitude')
    grid on
end

Answer 1

[Sreeram]

Hi Navid,

As per my understanding, you are trying to plot approximation (A) and detail (D) for all the sheets.

The code you provided uses subplot(1,2,p). This divides the current figure into an 1-by-2 grid and will place the plot in position specified by p. You may read more about the MATLAB function subplot here: https://www.mathworks.com/help/releases/R2023b/matlab/ref/subplot.html

To have the A and D plots for all the sheets in a single figure, you may want to choose number of rows in the grid of the figure to be number of sheets in the Excel file (num_sheets) and modify p accordingly.

Here’s how you can implement it:

clc
close all
clear all

filename = 's.xlsx';

% Get the sheet names
[~, sheetNames] = xlsfinfo(filename);

num_sheets = numel(sheetNames);
for i = 1:num_sheets
    % Read the data from the sheet
    data = xlsread(filename, i);
    % Extract the time and signal columns
    t(:,i) = data(:, 1);
    sig (:,i) = data(:, 2);
    N = length(sig);
    dt = t(3) - t(2); % sampling time
    fs = 1/dt; % freq
    signal = sig (:,i);
    wname = 'bior6.8';
    % Discrete Wavelet, DWT
    [CA1,CD1] = dwt(signal,wname);
    A1 = idwt(CA1,[],wname,N);
    D1 = idwt([],CD1,wname,N);

    t (:,i) = linspace(0,N,N)*(1/fs);
    subplot(num_sheets,2,2*i - 1);                  % Note the changes made to the subplot call
    plot (t,A1,'k','LineWidth',1.5);
    title(sprintf('Approximation for sheet%d', i));
    set(gca,'fontname','Times New Roman','FontSize',10)
    xlabel('Time (secs)')
    ylabel('Amplitude')
    grid on
    subplot(num_sheets,2,2*i);                      % Note the changes made to the subplot call
    plot (t,D1,'k','LineWidth',1.5);
    title(sprintf('Detail for sheet%d', i));
    set(gca,'fontname','Times New Roman','FontSize',10)
    xlabel('Time (secs)')
    ylabel('Amplitude')
    grid on
end
set(gcf,'position',[10,10,850,1600])                % So that the output below is legible

Hope it helps.