Cut a continuous signal into frames
3 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Hey everyone,
I have a continuous signal with size 1x400000 (analog data that corresponds to a 20 min signal) and I would like to frames with 1 min each and do some signal processing on each. I wrote this code and tried it for a small matrix but when I increased the size of my matrix matlab was busy for a very long time. So, what is wrong and what can I do to improve it.
The code: t= [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 17 17.01 17.02 17.03 17.4 18];
for i =1:length(t)
if t(i) <= 17
T(i)= t(i);
else
break;
end
end
plot(T, D(1:length(T))) %matrix D corresponds to the amplitude of my signal
when I used t= [1*ones(1,400000) 17 17 17.01 17.02 17.03 17.4 18] with the same code the program just like hanged!!!!
Any help will be appreciated.
4 comentarios
Respuesta aceptada
Wayne King
el 7 de Jul. de 2012
Editada: Wayne King
el 7 de Jul. de 2012
I'm guessing by your description, 400000 samples in 20 minutes that your sampling interval is 3 msec. So I'll use that value which gives 20000 samples in 1 minute.
You can just use reshape()
% just making up some data the length of your signal
x = randn(400000,1);
xnew = reshape(x,20000,20);
Now each column of the matrix xnew is a segment (frame) of your signal 1-minute in length, you can now do your processing on the columns of the matrix.
If you have the Signal Processing Toolbox, see the help for buffer() also, which enables you to do the same thing but works when you want some overlap between the columns, or you do not have exactly the number of samples required for reshape.
1 comentario
Más respuestas (0)
Ver también
Categorías
Más información sobre Get Started with Signal Processing Toolbox en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!