1
$\begingroup$

I have gigabytes worth of seismic ground data, that I wish to upsample by a factor of 2 or 3. I wanted to ask what efficient way might there be for me to upsample such a big file? Is there a way to break it down and do it in mini-files? (I am using MATLAB for my software, I think the resample command might take too long though). Thank you.

$\endgroup$
9
  • $\begingroup$ How good of resampling do you need? Simple linear interpolation between points can work if you don't need a high-quality resampler. $\endgroup$
    – Jason R
    Commented Jul 11, 2013 at 16:10
  • $\begingroup$ @JasonR I need a 'good enough' resampler. I am not sure how to be more specific on it though, or how to quantify that. Just as one might resample a vector by a factor of 2 or 3, so too, I need to resample this vector with the same expectations. $\endgroup$ Commented Jul 11, 2013 at 16:49
  • 1
    $\begingroup$ ''Good enough" will depend on the highest frequency components in your data set, and how accurately you want these frequencies reproduced or lack of aliased frequency content above them introduced. Also on the S/N ratio of the original data and the desired result. $\endgroup$
    – hotpaw2
    Commented Jul 11, 2013 at 17:43
  • 1
    $\begingroup$ I hope you mean: "...with a sampling rate of 100 KHz, I have some signals in the 45 KHz range that I want to maintain..." ? $\endgroup$
    – Paul R
    Commented Jul 11, 2013 at 22:10
  • 3
    $\begingroup$ Can you clarify why you want to upsample the data ? You don't gain any new information of course, and you double or treble your storage requirements. If you need upsampled data then why not just upsample on-the-fly when you read the data from disk ? $\endgroup$
    – Paul R
    Commented Jul 11, 2013 at 22:12

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.