I have a time series consisting of [time, value] pairs. The series is ordered by time but the difference between two consequent times is not constant. For example:

time, value
[0.1,  5.2]
[0.23, 6.9]
[0.3,  7.2]
[0.31, 4.8]

What are the methods to convert such time series to a vector with a constant sampling frequency? This will probably involve some interpolation of values that were nonexistent before, but that is ok for me.

Is there any matlab function for doing so?

The final goal is to be able to apply filters and FFT on the original timeseries.

  • $\begingroup$ not familiar with matlab, but you will need to do interpolation between the data. see se.mathworks.com/help/matlab/ref/interp1.html you may be able to define a strategy/algorithm that will produce a better result than a linear interpolation. $\endgroup$ – AlexanderBrevig Nov 19 '15 at 10:35
  • $\begingroup$ Is your signal band limited? $\endgroup$ – Moti Nov 21 '15 at 6:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.