def _agc_rms(data, window_size, desired_rms):
"""
Computes the RMS-amplitude AGC for data.
:param data: ``list`` of data values to compute rms for.
:param window_size: Length of window in number of samples.
:param desired_rms: Root-mean-squared value of ``data * _agc_rms(data)``.
:return: ``numpy.nparray`` with ``len(data)`` AGC values
"""
return float(desired_rms) / windowed_rms(data, window_size)
def test_windowed_rms(self):
"""
Should return the windowed RMS of a data array.
"""
rms0 = [ 1.0000000, 1.58113883, 2.54950976, 3.53553391, 4.52769257]
# should work on a list
a = [1, 2, 3, 4, 5]
rms1 = amplitudes.windowed_rms(a, 2)
for i in range(0, len(a)):
self.assertAlmostEqual(rms0[i], rms1[i])
# should work on a numpy.ndarray
a = np.asarray(a)
rms1 = amplitudes.windowed_rms(a, 2)
for i in range(0, len(a)):
self.assertAlmostEqual(rms0[i], rms1[i])
# rms of homogenous data should be the same
a0 = 20.32411
a = a0 * np.ones((100))
for w in [1, 10, 100]:
rms = amplitudes.windowed_rms(a, w)
for _rms in rms:
self.assertEqual(a0, _rms)
