Skip to contents

Filter signals in a PhysioExperiment object

Source: R/ops-filter.R
filterSignals.Rd

Applies a Butterworth IIR filter to signal data stored in a PhysioExperiment object. The filter is applied using zero-phase filtering (forward-backward) to avoid phase distortion.

Usage

filterSignals(
  pe,
  type = c("lowpass", "highpass", "bandpass", "bandstop"),
  cutoff,
  order = 4,
  assay_name = NULL,
  output_assay = NULL
)

Arguments

pe

A PhysioExperiment object.

type

Filter type: "lowpass", "highpass", "bandpass", or "bandstop".

cutoff

Cutoff frequency in Hz. A single value for "lowpass" or "highpass"; a length-2 vector c(low, high) for "bandpass" or "bandstop".

order

Filter order (default: 4).

assay_name

Which assay to filter. If NULL, uses the first assay.

output_assay

Name for the output assay. If NULL, defaults to "{assay_name}_filtered".

Value

A PhysioExperiment object with filtered data stored as a new assay.

Details

This function requires the signal package (listed in Suggests). If not installed, an informative error message with install instructions is provided.

The Butterworth filter is designed using signal::butter() and applied with signal::filtfilt() for zero-phase filtering. The cutoff frequency is normalized to the Nyquist frequency automatically.

References

Winter DA (2009). "Biomechanics and Motor Control of Human Movement." 4th ed. John Wiley & Sons.

Butterworth S (1930). "On the Theory of Filter Amplifiers." Wireless Engineer, 7, 536-541.

See also

butterworthFilter() for filtering raw vectors and matrices, savgolFilter() for Savitzky-Golay polynomial smoothing, movingAverage() for simple moving average smoothing.

Examples

if (FALSE) { # \dontrun{
pe <- make_mocap_markers(n_time = 500, n_markers = 4, sr = 120)
pe_filt <- filterSignals(pe, type = "lowpass", cutoff = 10)
} # }