Signals

arora.signals.fourier

arora.signals.fourier(epochized_data, sampling_frequency)

Apply fourier fast transform method to a list of signals using fft() from SciPy.fftpack.

Parameters:

signalslist of list of int or float

The segmented data in a list

sampling_frequencyint or float

The sampling frequency of the data

Returns:

List of transformed data where each epoch is a NumPy array.

For example:

>>> some_data = [[1, 2], [2, 3], [3, 4]]
>>> arora.signals.fourier(some_data, 200)
[array([ 3.-0.j, -1.-0.j]), array([ 5.-0.j, -1.-0.j]), array([ 7.-0.j, -1.-0.j])]

arora.signals.resample

arora.signals.resample(signals, old_fs, new_fs)

Resamples the signals given both the new frequency and the old frequency.

Parameters:

signalslist of int or float

List of signals.

old_fsint

The old frequency of the signal

new_fsint

The new frequency of the signal

Returns :

NumPy array of resampled data.

For example:

>>> signals = [1, 2, 3, 4, 5, 6, 7, 8]
>>> arora.signals.resample(signals, 4, 1)
array([1.        , 0.32651433, 0.47266051, 1.16309138, 2.    ,
   2.64941807, 2.96926627, 3.02470348, 3.        , 3.06911053,
   3.30108763, 3.64674704, 4.        , 4.28360902, 4.5       ,
   4.71639098, 5.        , 5.35325296, 5.69891237, 5.93088947,
   6.        , 5.97529652, 6.03073373, 6.35058193, 7.        ,
   7.83690862, 8.52733949, 8.67348567, 8.        , 6.50588996,
   4.5       , 2.49411004])

arora.signals.lower_frequency

arora.signals.lower_frequency(signals, method, new_fs)

Lowers the frequency of the signals with given method. Uses arora.signals.segment.fs() to segment the signals with given frequency.

Parameters:

signals: list of int or float

List of time series signals.

method: {‘min’ or ‘max’ or ‘mean’ or ‘median’}, default ‘mean’

The method how the frequency should be lowered.

new_fs: int

The new frequency of the signals.

Returns:

List of new values corresponding to the method given. Length of returned list = len(signals) / new_fs.

For example:

>>> signals = [1,2,3,4,5,6,7,8]
>>> arora.signals.lower_frequency(signals, 2, min)
[1, 3, 5, 7]
>>> arora.signals.lower_frequency(signals, 2, max)
[2, 4, 6, 8]
>>> arora.signals.lower_frequency(signals, 2, mean)
[1.5, 3.5, 5.5, 7.5]
>>> arora.signals.lower_frequency(signals, 4, median)
[2.5, 6.5]

arora.math.welch_psd

arora.math.welch_psd(signals, fs, filter_order)

Estimate power spectral density using Welch’s method Smoothing

Parameters:

signalspd.DataFrame

An array of a signal

sampling_frequencyint or float

The frequency of the epoch

filter_orderint

Order of the filter

Returns:

Arrays of sample frequency as numpy.ndarray.