Array prealocation - Extra array of zeros in 3D array

1 Ansicht (letzte 30 Tage)
Hi,
I have a 13 channel signal recorded over several hours. I wish to take all channels, split them into 10 second windows with a 5 second overlap and finally concatinate all channels into a 3D array (Buffer length(n), number of buffers, channels).
I have written the function below which works as intended, except my preallocate method results in an 'extra channel' of zeros, 14 rather than 13 with the 1st array containing zeros. What I think would work is windows3D = zeros(size(windows)), however I cannot call that as 'window' is generated after the pre-allocation call.
I could just go into the array after the fact and 'delete' the additonal array of zeros however this feels like cheating, is there a better way to do this?
Kind regards,
Christopher
function [windows3D] = windowGen(eegDataNorm)
% 10 second window, 50% overlap
EEG = eegDataNorm; % Short label
Fs = 200; % Sampling frequency (Hz)
windows3D = zeros(2000,5147); %Preallocate
for i = 1:size(eegDataNorm)
windows = buffer(eegDataNorm(i,:),10*Fs,5*Fs);
windows3D = cat(3,windows3D,windows);
end
end

Akzeptierte Antwort

Stephen23
Stephen23 am 2 Nov. 2021
Bearbeitet: Stephen23 am 2 Nov. 2021
windows3D = nan(2000,5147,0);
or use actual preallocation:
N = size(eegDataNorm,1);
windows3D = zeros(2000,5147,N);
for k = 1:N
windows3D(:,:,k) = buffer(..)
end
  1 Kommentar
Christopher McCausland
Christopher McCausland am 2 Nov. 2021
Hi Stephen,
I just wanted to say thank you for you answer, I think I may have been looking at my code for too long... not sure how I didn't see this myself! For anyone else here's what the final fuction looks like. I also added a slight modifcation to allow diffrent lenght data to be used which works nicely.
function [windows3D] = windowGen(eegDataNorm)
% 10 second window, 50% overlap
EEG = eegDataNorm; % Short label
Fs = 200; % Sampling frequency (Hz)
C = size(eegDataNorm,1);
[N,P] = size(buffer(eegDataNorm(1,:),10*Fs,5*Fs));
windows3D = zeros(N,P,C); %Prealocate
for i = 1:C
windows3D(:,:,i) = buffer(eegDataNorm(i,:),10*Fs,5*Fs);
end
end

Melden Sie sich an, um zu kommentieren.

Weitere Antworten (0)

Kategorien

Mehr zu EEG/MEG/ECoG finden Sie in Help Center und File Exchange

Produkte


Version

R2020b

Community Treasure Hunt

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

Start Hunting!

Translated by