(Not recommended) Resample time-domain data that is stored in an
iddata object by decimation or interpolation (requires
Signal Processing Toolbox software)
resample is not recommended. Use
idresamp instead to resample estimation data in the form of timetables,
numeric matrices, and
iddata objects, with or without
Signal Processing Toolbox™. For more information, see Compatibility Considerations.
Resample Time-Domain Data
Increase the sampling rate of data and compare the resampled and the original data signals.
idinput to generate an input signal
u that contains 2 periods, each containing 20 samples, of a sum-of-sinusoids signal with 5 sinusoids.
u = idinput([20 1 2],'sine',,,[5 10 1]); plot(u)
u into an
ui. Set the start time to
0. A sum-of-sinusoids signal is band limited, so set
ui = iddata(,u,1,'tstart',0,'intersample','bl');
Resample the data using factors of 25 and 3 for
Q, respectively. These values produce an increase in sample rate of about 8.3.
ur = resample(ui,25,3); plot(ui,ur) legend('ui','ur')
The resampled signal is smoother than the original signal.
data — Original data
Original data, specified as an
iddata that contains either
input/output data or time series data.
contain uniform time samples.
P,Q — Resampling factor terms
Resampling factor terms, specified as integers that represent the
interpolation and decimation factors. After resampling, the new sample time
Q/P times the original sample time.
(Q/P)>1 results in decimation and setting
(Q/P)<1 results in interpolation.
order — Order of FIR filter
10 (default) | positive integer
Order of the FIR filter that
resample applies before
interpolation and decimation, specified as a positive integer.
resample calls the
resample (Signal Processing Toolbox) function in Signal Processing Toolbox. The algorithm takes into account the intersample characteristics of the
input signal that are described by the
If you do not have a license for Signal Processing Toolbox, perform resampling using