- why filter before resampling?
- what kind of filter are u using?
Reduce time with this
2 views (last 30 days)
Show older comments
hi all!
i've a filter that is used on the data, but there is lots of data! i am hoping to resample the data(ia take every 2nd point), and than filter it. the problem is that i only have half as many data points. how can i use spline on the filtered data to make it back to the original.
for example, say theres 1000 data points. i resample every other data point to have 500. i feed that data into filter, and have newdata. how can i use spline to fit back to 1000 data points? i tried resampling every 1/2 but that is no good.
any suggestions will be helpful
thank you!!
1 Comment
Azzi Abdelmalek
on 19 Aug 2012
Edited: Azzi Abdelmalek
on 19 Aug 2012
if you are using a filter to just remove eventual noise; and don't affect considerably your initial signal; then you have just to interpolate your 500 data. just respect Shanon theorem for sampling
Answers (3)
Azzi Abdelmalek
on 19 Aug 2012
Edited: Azzi Abdelmalek
on 19 Aug 2012
example
x=0.2:0.02:10;y=sin(x) %signal with 500 points % sig
xi=0.01:0.01:10; % 1000 points
yi=interp1(x,y,xi,'spline') % interpolated signal yi for points xi
0 Comments
Jan
on 19 Aug 2012
It is a bad idea to omit half of the data before filtering, because the re-creation of these points by using a spline interpolation creates(!) noise again. In addition the interpolation and will take much more time than filtering the signal in its original rate. If the filtering needs too much time for you, try the faster C-Mex FEX: FilterM.
See Also
Categories
Find more on Multirate Signal Processing in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!