Error using fft on code

% Not sure what I am doing wrong to be getting an error with the data type

% Load data from Signal.mat file
load('Spec.mat');
% Number of rows (process realizations)
num_rows = size(Spec, 1);
% Number of samples per row
num_samples = size(Spec, 2);
% Calculate the power spectral density (PSD) for each row
psd_rows = zeros(num_rows, num_samples);
for i = 1:num_rows
    psd_rows(i, :) = (abs(fftshift(fft(Spec(i, :)))).^2) / num_samples; <------------- error here
end
% Average the PSDs of all rows
psd_average = mean(psd_rows, 1);
% Sampling frequency (assuming 40 kHz)
Fs = 40e3;
% Frequency resolution
df = Fs / num_samples;
% Frequency axis
f_axis = (-num_samples/2 : num_samples/2 - 1) * df;
% Plot individual and composite PSDs
figure;
hold on;
for i = 1:num_rows
    plot(f_axis, fftshift(psd_rows(i, :)));
end
plot(f_axis, fftshift(psd_average), 'k', 'LineWidth', 2);
xlabel('Frequency (Hz)');
ylabel('Power/Frequency');
title('Power Spectral Density (PSD)');
legend('Individual PSDs', 'Composite PSD');

[Voss]

What is the error message?

It will probably also help to have the file Spec.mat. You can upload it using the paperclip button.

Answer 1

[Gayatri]

Hi Matthew,

When you load a MAT file using "load" function, assign it to a variable. Data inside the structure can be accessed as below :

% Load data from Spec.mat file
Data = load('Spec.mat');
% Number of rows (process realizations)
num_rows = size(Data.Spec, 1);

I have created the sample Spec MAT file for testing purpose and not getting any error on given line:

% Number of signals (rows)
num_signals = 5;
% Number of samples (columns)
num_samples = 1024;
% Generate random data for simplicity
Spec = randn(num_signals, num_samples);
% Save the matrix to a .mat file
save('Spec.mat', 'Spec');

% Load data from Spec.mat file
Data = load('Spec.mat');
% Number of rows (process realizations)
num_rows = size(Data.Spec, 1);
% Number of samples per row
num_samples = size(Data.Spec, 2); % Corrected from 'size(Spec, 2);' to 'size(Data.Spec, 2);'
% Calculate the power spectral density (PSD) for each row
psd_rows = zeros(num_rows, num_samples);
for i = 1:num_rows
    psd_rows(i, :) = (abs(fftshift(fft(Data.Spec(i, :)))).^2) / num_samples; 
end
% Average the PSDs of all rows
psd_average = mean(psd_rows, 1);
% Sampling frequency (assuming 40 kHz)
Fs = 40e3;
% Frequency resolution
df = Fs / num_samples;
% Frequency axis
f_axis = (-num_samples/2 : num_samples/2 - 1) * df;
% Plot individual and composite PSDs
figure;
hold on;
for i = 1:num_rows
    plot(f_axis, fftshift(psd_rows(i, :)));
end
plot(f_axis, fftshift(psd_average), 'k', 'LineWidth', 2);
xlabel('Frequency (Hz)');
ylabel('Power/Frequency');
title('Power Spectral Density (PSD)');
legend('Individual PSDs', 'Composite PSD');

I hope it helps!