MATLAB Answers

Todd
0

I have a rather large matrix. The data was sampled at 100 Hz. I want to the data to look like it was sampled at 5 Hz

Asked by Todd
on 17 Aug 2012
Right now the left column is time and it counts .01, .02, .03 etc.
I want to pull out all of the data at 5 hz intervals such as 0, .2, .4 and delete the rest of the rows.
I will answer any question for clarification too

  4 Comments

Show 1 older comment
by the way , I don't know what the goal is but if you have been oversampling maybe it is better to take the mean of the moving average, depending on your data averaging will probaly decrease noise
yeah all right all depends on what you are measuring and the first thing to do is to analyse your data in a smart way of course, but let us say that you are doing e.g. loadcell measurements over time to see how a the force changes, you will have noise on the measurement and the average will gie you an idea how the signal moves in time , that is what I meant

Sign in to comment.

Tags

4 Answers

Answer by Jürgen on 17 Aug 2012

well If I understand it well you have a vector Sample100 with 100 Hz Samples and you want only the sample @ 5 hz Sample5=Sample100(1:20:end) should do the trick then hope this helps regards,J

  0 Comments

Sign in to comment.


Answer by Azzi Abdelmalek
on 17 Aug 2012
Edited by Azzi Abdelmalek
on 17 Aug 2012

fe=100 %sample frequency
te=1/fe %sample time
t=0:te:1;y=rand(length(t),1)%
fd=5 %desired frequency
td=1/fd %desired sample time
nt=floor(td/te)
ind=1:nt:length(t);
new_t=t(ind);
new_y=y(ind);
plot(t,y);hold on;plot(new_t,new_y,'xr','linewidth',14);

  0 Comments

Sign in to comment.


Answer by Loginatorist on 17 Aug 2012

Here is another example. You say the first column is time, so it is here.
% Original data with columns: [time,signal]
S = [(0:.01:2*pi).' sin(0:.01:2*pi).'];
Sd = S(1:20:end,:); % Downsampled signal.
plot(S(:,1),S(:,2),'b.',Sd(:,1),Sd(:,2),'sr')
legend('Original','Downsampled')

  0 Comments

Sign in to comment.


Answer by Dr. Seis
on 17 Aug 2012
Edited by Dr. Seis
on 17 Aug 2012

Use resample - This is an example of why you may not want to take every Nth amplitude:
Example where a is your input dataset and b is your output dataset:
Fs_in = 100; % samples per second
Fs_out = 5; % samples per second
b = resample(a,Fs_out,Fs_in);

  0 Comments

Sign in to comment.