How to extract frequency sub-bands of an EEG signal using wavelet packet transform?

42 views (last 30 days)
i want to decompose a signal with wavelet in 5 subbands: delta (0–4 Hz), theta (4–8 Hz), alpha (8–15 Hz), beta (15–30 Hz), and gamma (30–60 Hz) the technical specifications of the signal sre as follows: "Signals are band-pass filtered between 0.05 and 200 Hz and then digitized at 1000 Hz with 16 bit (0.1 uV) accuracy." So kindly suggest me how can i extract these sub-bands using Wavelet Packet Decomposition. i have tried a lot but could not do the same. suggest me the relevant commands for this purpose.
  2 Comments
Rasa
Rasa on 5 Apr 2019
Hello, Did you get any solution on the problem ? I also want to know how to do this. Please kindly share.
Thanks
kalarmago
kalarmago on 17 Apr 2020
Hi, I am in the same issue, please, someone could help me and put some code (I am new in Matlab)?

Sign in to comment.

Answers (1)

Abdulhakim Alezzi
Abdulhakim Alezzi on 18 Sep 2020
load eegdata.mat;
s=eegdata;
figure;p=plot(s);
title('EEG Signal')
fs = 500;
% Sampling frequency
N=length(s);
waveletFunction = 'db8';
[C,L] = wavedec(s,8,waveletFunction);
cD1 = detcoef(C,L,1);
cD2 = detcoef(C,L,2);
cD3 = detcoef(C,L,3);
cD4 = detcoef(C,L,4);
cD5 = detcoef(C,L,5); %GAMA
cD6 = detcoef(C,L,6); %BETA
cD7 = detcoef(C,L,7); %ALPHA
cD8 = detcoef(C,L,8); %THETA
cA8 = appcoef(C,L,waveletFunction,8); %DELTA
D1 = wrcoef('d',C,L,waveletFunction,1);
D2 = wrcoef('d',C,L,waveletFunction,2);
D3 = wrcoef('d',C,L,waveletFunction,3);
D4 = wrcoef('d',C,L,waveletFunction,4);
D5 = wrcoef('d',C,L,waveletFunction,5); %GAMMA
D6 = wrcoef('d',C,L,waveletFunction,6); %BETA
D7 = wrcoef('d',C,L,waveletFunction,7); %ALPHA
D8 = wrcoef('d',C,L,waveletFunction,8); %THETA
A8 = wrcoef('a',C,L,waveletFunction,8); %DELTA
Gamma = D5;
figure; subplot(5,1,1); plot(1:1:length(Gamma),Gamma);title('GAMMA');
Beta = D6;
subplot(5,1,2); plot(1:1:length(Beta), Beta); title('BETA');
Alpha = D7;
subplot(5,1,3); plot(1:1:length(Alpha),Alpha); title('ALPHA');
Theta = D8;
subplot(5,1,4); plot(1:1:length(Theta),Theta);title('THETA');
D8 = detrend(D8,0);
Delta = A8;
%figure, plot(0:1/fs:1,Delta);
subplot(5,1,5);plot(1:1:length(Delta),Delta);title('DELTA');
D5 = detrend(D5,0);
xdft = fft(D5);
freq = 0:N/length(D5):N/2;
xdft = xdft(1:length(D5)/2+1);
figure;subplot(511);plot(freq,abs(xdft));title('GAMMA-FREQUENCY');
[~,I] = max(abs(xdft));
fprintf('Gamma:Maximum occurs at %3.2f Hz.\n',freq(I));
D6 = detrend(D6,0);
xdft2 = fft(D6);
freq2 = 0:N/length(D6):N/2;
xdft2 = xdft2(1:length(D6)/2+1);
% figure;
subplot(512);plot(freq2,abs(xdft2));title('BETA');
[~,I] = max(abs(xdft2));
fprintf('Beta:Maximum occurs at %3.2f Hz.\n',freq2(I));
D7 = detrend(D7,0);
xdft3 = fft(D7);
freq3 = 0:N/length(D7):N/2;
xdft3 = xdft3(1:length(D7)/2+1);
% figure;
subplot(513);plot(freq3,abs(xdft3));title('ALPHA');
[~,I] = max(abs(xdft3));
fprintf('Alpha:Maximum occurs at %f Hz.\n',freq3(I));
xdft4 = fft(D8);
freq4 = 0:N/length(D8):N/2;
xdft4 = xdft4(1:length(D8)/2+1);
% figure;
subplot(514);plot(freq4,abs(xdft4));title('THETA');
[~,I] = max(abs(xdft4));
fprintf('Theta:Maximum occurs at %f Hz.\n',freq4(I));
A8 = detrend(A8,0);
xdft5 = fft(A8);
freq5 = 0:N/length(A8):N/2;
xdft5 = xdft5(1:length(A8)/2+1);
% figure;
subplot(515);plot(freq3,abs(xdft5));title('DELTA');
[~,I] = max(abs(xdft5));
fprintf('Delta:Maximum occurs at %f Hz.\n',freq5(I));
  1 Comment
Piyush Kant
Piyush Kant on 29 Sep 2020
can you please explain the level selection for the same as Fs=500 thus making max fequency content 0-250Hz so according to me
d1=125-250 Hz
d2= 62.5-125 Hz
d3= 31.75-62.5Hz
d4 = 15.87 - 31.75Hz
d5 ~= 8-16 Hz
d6~= 4-8 Hz
d7=2-4 Hz
d8 = 1-2 Hz
Please correct me if i am wrong

Sign in to comment.

Categories

Find more on EEG/MEG/ECoG in Help Center and File Exchange

Products

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!