import numpy
import scipy.stats
[docs]
def harmonic(f, t):
"""
Returns a sine wave of the specified frequency.
Parameters
----------
f : float
Signal frequency in hertz.
t : array
Timeline array.
"""
return numpy.sin(2 * numpy.pi * f * t)
[docs]
def harmonics(f, n, t):
"""
Returns a harmonic wave including the specified amount of overtones.
Parameters
----------
f : float
Fundamental signal frequency in hertz.
n : int
Number of additional overtones.
t : array
Timeline array.
"""
assert isinstance(n, int)
n = numpy.abs(n)
m = numpy.arange(1, n + 2)
a = 1 / m
o = [a[i - 1] * harmonic(f * i, t) for i in m]
return numpy.stack(o).sum(axis=0) / numpy.sum(a)
[docs]
def square(f, t):
"""
Returns a square shaped wave of the specified frequency.
Parameters
----------
f : float
Signal frequency in hertz.
t : array
Timeline array.
"""
return numpy.sign(harmonic(f, t))
[docs]
def sawtooth(f, t):
"""
Returns a sawtooth shaped wave of the specified frequency.
Parameters
----------
f : float
Signal frequency in hertz.
t : array
Timeline array.
"""
return 2 * (f * t - numpy.floor(f * t + 0.5))
[docs]
def triangle(f, t):
"""
Returns a triangle shaped wave of the specified frequency.
Parameters
----------
f : float
Signal frequency in hertz.
t : array
Timeline array.
"""
return 2 * numpy.abs(sawtooth(f, t)) - 1
[docs]
def chirp(f0, f1, t):
"""
Returns a chirp signal in which the frequency alterates over the time.
Parameters
----------
f0 : float
Start signal frequency in hertz.
f1 : float
End signal frequency in hertz.
t : array
Timeline array.
"""
return numpy.sin(2 * numpy.pi * ((t ** 2) * (f1 - f0) / 2) + t * f0)
[docs]
def noise(t, s=1):
"""
Returns a random white noise based on the truncated normal distribution.
Parameters
----------
t : array
Timeline array.
s : float, optional
Scale parameter between 0 and 1.
"""
a = -1 / s
b = +1 / s
n = len(t)
return scipy.stats.truncnorm(a, b, loc=0, scale=s).rvs(n)