Main Content

idealfilter

timeseries ideal filter

Description

example

tsout = idealfilter(tsin,interval,filtertype) applies an ideal (noncausal) filter of type filtertype to the frequency intervals specified by interval for a timeseries object tsin.

Ideal filters are noncausal, and the ends of the filter amplitude are flat in the frequency domain. The data in ts must have zero mean.

tsout = idealfilter(tsin,interval,filtertype,ind) optionally specifies the row or column indices of tsin to apply the filter to.

Examples

collapse all

First apply an ideal notch filter to a timeseries object, then apply a pass filter.

Load the data in the file count.dat, and create a timeseries object from the matrix count.

load count.dat
tsin = timeseries(count(:,1),1:24);

Compute the mean of the data in tsin.

tsinmean = mean(tsin);

Define the frequency interval, in hertz, for filtering the data.

interval = [0.08 0.2];

Invoke an ideal notch filter.

tsoutnotch = idealfilter(tsin,interval,'notch');

Compare the original data and the filtered data.

plot(tsin,'-.')
hold on
plot(tsoutnotch,'-')

Restore the mean to the filtered data.

tsoutnotchmean = tsoutnotch + tsinmean;
plot(tsoutnotchmean,':')
title('Notch Filter')
legend('Original Data','Filtered Data','Mean Restored',...
       'Location','NorthWest')
hold off 

Figure contains an axes object. The axes object with title Notch Filter, xlabel Time (seconds), ylabel unnamed contains 3 objects of type line. These objects represent Original Data, Filtered Data, Mean Restored.

Repeat the filtering process using a pass filter.

plot(tsin,'-.')
hold on
tsoutpass = idealfilter(tsin,interval,'pass');
plot(tsoutpass,'-')

tsoutpassmean = tsoutpass + tsinmean;
plot(tsoutpassmean,':')
title('Pass Filter')
legend('Original Data','Filtered Data','Mean Restored',...
       'Location','NorthWest')

Figure contains an axes object. The axes object with title Pass Filter, xlabel Time (seconds), ylabel unnamed contains 3 objects of type line. These objects represent Original Data, Filtered Data, Mean Restored.

Input Arguments

collapse all

Input timeseries with zero mean, specified as a scalar.

  • If tsin is nonuniformly sampled, then idealfilter resamples the data on a uniform time vector before applying the filter.

  • idealfilter replaces any NaN elements of tsin using the interpolation method associated with tsin prior to applying the filter.

Data Types: timeseries

Frequency interval, specified as a two-column matrix where each row represents the start and end frequencies for each interval.

Data Types: double | single | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

Filter type, specified as one of the following options:

  • 'pass' — Allow variations in a specific frequency range

  • 'notch' — Remove variations in a specific frequency range

Data Types: double | single | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

Row or column indices, specified as a positive integer numeric scalar or vector. ind represents column indices for column-oriented data (tsin.IsTimeFirst is true) and represents row indices for row-oriented data (tsin.IsTimeFirst is false).

Data Types: double | single | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

Version History

Introduced before R2006a

See Also

|