I apply SWT method to remove artifact from EEG signal. The level of decomposition,5 and the length of the signal,2000.How to apply 2^Level in this code? What can I do?

DataSet

%%

clc;

clear

close all;

load B02T.mat;

fs=250;

% t = 0.004:1/fs:1;

eeg1 = data{1,1}.X;

channel_1= eeg1(:,1);

ch_1=channel_1(1:2000);

figure;

subplot(2,1,1);

plot(ch_1);

xlabel('Time (s)');

ylabel('Amplitude');

legend('ch1');

title("Plot channel "+1+ " for Structure 1X1");

hold on;

grid on;

eeg_signal = eeg1(1:2000); % first 2000 samples

subplot(2,1,2);

plot(eeg_signal,'black');

hold on;

grid on;

legend('EEG Signal');

title('Raw EEG Signal');

%%

% Set parameters

wavelet = 'db4'; % Wavelet family

level = 5; % Level of decomposition

% Perform the SWT on the raw signal

[swt_coeffs, ~] = swt(eeg_signal, level, wavelet);

% Plot the subbands of the SWT coefficients

figure;

for i = 1:level+1

subplot(level+1, 1, i);

plot(swt_coeffs(i, :));

title(['Subband ', num2str(i)]);

xlabel('Sample');

ylabel('Amplitude');

end

% Set the global threshold using a threshold selection rule or a fixed threshold value

thr = wthrmngr('sw2ddenoLVL', 'penalhi', swt_coeffs);

% Create an array to store denoised subbands

denoised_subbands = cell(level+1, 1);

% Apply thresholding to each subband

for i = 1:level

% Extract coefficients for the current subband

subband_coeffs = swt_coeffs(i, :);

% Apply global thresholding to the subband coefficients

denoised_subband_coeffs = subband_coeffs;

denoised_subband_coeffs(abs(denoised_subband_coeffs) < thr) = 0;

% Store denoised subband coefficients

denoised_subbands{i} = denoised_subband_coeffs;

end

% Reconstruct the denoised EEG signal by combining the denoised subbands

denoised_coeffs = [swt_coeffs(level+1, :), denoised_subbands{level:-1:1}];

cleaned_signal = iswt(denoised_coeffs, wavelet);

noise_signal = eeg_signal - cleaned_signal;

% Plot the original and denoised signals

t = 1:length(eeg_signal);

figure;

subplot(3, 1, 1);

plot(t, eeg_signal);

title('Original EEG Signal');

xlabel('Time');

ylabel('Amplitude');

subplot(3, 1, 2);

plot(t, cleaned_signal);

title('Denoised EEG Signal');

xlabel('Time');

ylabel('Amplitude');

subplot(3, 1, 3);

plot(t, noise_signal);

title('noised EEG Signal');

xlabel('Time');

ylabel('Amplitude');

%%

% Calculate SNR

SNR = 20 * log10(norm(eeg_signal)/ norm(noise_signal));

% Calculate MSE

MSE = ((norm(eeg_signal) - norm(cleaned_signal)).^2) / 2000;

% MSE = immse(norm(eeg_signal), norm(cleaned_signal))/2000;

disp(['SNR: ', num2str(SNR), ' dB']);

disp(['MSE: ', num2str(MSE)]);

Answer 1

[Diwakar Diwakar]

To apply the `2^Level` in the code, you can modify the following lines:

level = 5; % Level of decomposition

Replace it with:

level = 5;

N = 2^level;

Then, when performing the SWT, use `N` as the length of the signal:

[swt_coeffs, ~] = swt(eeg_signal, N, wavelet);

Similarly, when storing the denoised subbands, use `N` as the length:

denoised_subbands = cell(N+1, 1);

In the loop for applying thresholding to each subband, change the loop condition:

for i = 1:N

Finally, when reconstructing the denoised EEG signal, modify the concatenation of coefficients:

denoised_coeffs = [swt_coeffs(N+1, :), denoised_subbands{N:-1:1}];

This way, you are using the desired level of decomposition, `level`, and applying `2^level` in the relevant parts of the code.

[Komal]

Thank you so much.