Added methods to filter bank designer to create narrow-band spectra from octave band results, assuming a uniform power spectrum in a certain band

This commit is contained in:
Anne de Jong 2020-01-13 14:43:25 +01:00
parent 2270a297cc
commit 8c6e6a5828

View File

@ -52,7 +52,6 @@ class FilterBankDesigner:
Returns:
True if filter is norm-compliant, False if not
"""
# Skip zero-frequency
if np.isclose(freq[0], 0):
@ -62,7 +61,8 @@ class FilterBankDesigner:
# Interpolate limites to frequency array as given
llim_full = np.interp(freq, freqlim, llim, left=-np.inf, right=-np.inf)
ulim_full = np.interp(freq, freqlim, ulim, left=ulim[0], right=ulim[-1])
ulim_full = np.interp(freq, freqlim, ulim,
left=ulim[0], right=ulim[-1])
return bool(np.all(llim_full <= h_dB) and
np.all(ulim_full >= h_dB))
@ -161,6 +161,66 @@ class FilterBankDesigner:
raise ValueError(
f'Could not find an x-value corresponding to {nom_txt}.')
def getxs(self, nom_txt_start, nom_txt_end):
"""Returns a list of all filter designators, for given start end end
nominal frequencies.
Args:
nom_txt_start: Start frequency band, i.e. '31.5'
nom_txt_end: End frequency band, i.e. '10k'
Returns:
[x0, x1, ..]
"""
xstart = self.nominal_txt_tox(nom_txt_start)
xend = self.nominal_txt_tox(nom_txt_end)
return list(range(xstart, xend+1))
def getNarrowBandFromOctaveBand(self, xl, xu,
levels_in_bands, npoints=500):
"""Create a narrow band spectrum based on a spectrum in (fractional)
octave bands. The result is create such that the total energy in each
frequency band is constant. The latter can be checked by working it
back to (fractional) octave bands, which is doen using the function
`getOctaveBandsFromNarrowBand`. Note that the resulting narrow band has
units of *power*, not power spectral density. Input should be levels in
**deciBells**.
Args:
xl: Band designator of lowest band
xu: Band designator of highest band
levels_in_bands: levels in dB for each band, should have length
(xu+1 - xl)
npoints: Number of discrete frequency points in linear frequency
array.
Returns:
freq, levels_dB. Where levels_dB is an array of narrow band levels
"""
# Lowest frequency of all frequencies
fll = self.fl(xl)
# Highest frequency of all frequencies
fuu = self.fu(xu)
freq = np.linspace(fll, fuu, npoints)
levels_narrow = np.empty_like(freq)
for i, x in enumerate(range(xl, xu + 1)):
fl = self.fl(x)
fu = self.fu(x)
# Find the indices in the frequency array which correspond to the
# frequency band x
if x != xu:
indices_cur = np.where((freq >= fl) & (freq < fu))
else:
indices_cur = np.where((freq >= fl) & (freq <= fu))
power_cur = 10**(levels_in_bands[i] / 10)
power_narrow = power_cur / indices_cur[0].size
level_narrow = 10*np.log10(power_narrow)
levels_narrow[indices_cur] = level_narrow
return freq, levels_narrow
class OctaveBankDesigner(FilterBankDesigner):
"""Octave band filter designer."""
@ -204,9 +264,11 @@ class OctaveBankDesigner(FilterBankDesigner):
mininf = -1e300
if filter_class == 1:
lower_limits_pos = [-0.3, -0.4, -0.6, -1.3, -5.0, -5.0] + 4*[mininf]
lower_limits_pos = [-0.3, -0.4, -
0.6, -1.3, -5.0, -5.0] + 4*[mininf]
elif filter_class == 0:
lower_limits_pos = [-0.15, -0.2, -0.4, -1.1, -4.5, -4.5] + 4*[mininf]
lower_limits_pos = [-0.15, -0.2, -
0.4, -1.1, -4.5, -4.5] + 4*[mininf]
lower_limits_neg = lower_limits_pos[:]
lower_limits_neg.reverse()
lower_limits = np.asarray(lower_limits_neg[:-1] + lower_limits_pos)
@ -223,7 +285,6 @@ class OctaveBankDesigner(FilterBankDesigner):
return freqs, lower_limits, upper_limits
def nominal_txt(self, x):
"""Returns textual repressentation of corresponding to the nominal
frequency."""