def myzpk2tf(self, z, p, k):
z = np.atleast_1d(z)
k = np.atleast_1d(k)
if len(z.shape) > 1:
temp = np.poly(z[0])
b = np.zeros((z.shape[0], z.shape[1] + 1), temp.dtype.char)
if len(k) == 1:
k = [k[0]] * z.shape[0]
for i in range(z.shape[0]):
b[i] = k[i] * poly(z[i])
else:
b = k * np.poly(z)
a = np.atleast_1d(np.poly(p))
# Use real output if possible. Copied from numpy.poly, since
# we can't depend on a specific version of numpy.
if issubclass(b.dtype.type, np.complexfloating):
# if complex roots are all complex conjugates, the roots are real.
roots = np.asarray(z, complex)
pos_roots = np.compress(roots.imag > 0, roots)
neg_roots = np.conjugate(np.compress(roots.imag < 0, roots))
if len(pos_roots) == len(neg_roots):
if np.all(np.sort_complex(neg_roots) == np.sort_complex(pos_roots)):
b = b.real.copy()
if issubclass(a.dtype.type, np.complexfloating):
# if complex roots are all complex conjugates, the roots are real.
roots = np.asarray(p, complex)
pos_roots = np.compress(roots.imag > 0, roots)
neg_roots = np.conjugate(np.compress(roots.imag < 0, roots))
if len(pos_roots) == len(neg_roots):
if np.all(np.sort_complex(neg_roots) == np.sort_complex(pos_roots)):
a = a.real.copy()
return b, a
def zpk2tf(z, p, k):
"""Return polynomial transfer function representation from zeros
and poles
Parameters
----------
z : ndarray
Zeros of the transfer function.
p : ndarray
Poles of the transfer function.
k : float
System gain.
Returns
-------
b : ndarray
Numerator polynomial.
a : ndarray
Denominator polynomial.
"""
z = atleast_1d(z)
k = atleast_1d(k)
if len(z.shape) > 1:
temp = poly(z[0])
b = zeros((z.shape[0], z.shape[1] + 1), temp.dtype.char)
if len(k) == 1:
k = [k[0]] * z.shape[0]
for i in range(z.shape[0]):
b[i] = k[i] * poly(z[i])
else:
b = k * poly(z)
a = atleast_1d(poly(p))
return b, a
def lsf2poly(lsf):
"""Convert line spectral frequencies to prediction filter coefficients
returns a vector a containing the prediction filter coefficients from a vector lsf of line spectral frequencies.
.. doctest::
>>> from spectrum import lsf2poly
>>> lsf = [0.7842 , 1.5605 , 1.8776 , 1.8984, 2.3593]
>>> a = lsf2poly(lsf)
# array([ 1.00000000e+00, 6.14837835e-01, 9.89884967e-01,
# 9.31594056e-05, 3.13713832e-03, -8.12002261e-03 ])
.. seealso:: poly2lsf, rc2poly, ac2poly, rc2is
"""
# Reference: A.M. Kondoz, "Digital Speech: Coding for Low Bit Rate Communications
# Systems" John Wiley & Sons 1994 ,Chapter 4
# Line spectral frequencies must be real.
lsf = numpy.array(lsf)
if max(lsf) > numpy.pi or min(lsf) < 0:
raise ValueError('Line spectral frequencies must be between 0 and pi.')
p = len(lsf) # model order
# Form zeros using the LSFs and unit amplitudes
z = numpy.exp(1.j * lsf)
# Separate the zeros to those belonging to P and Q
rQ = z[0::2]
rP = z[1::2]
# Include the conjugates as well
rQ = numpy.concatenate((rQ, rQ.conjugate()))
rP = numpy.concatenate((rP, rP.conjugate()))
# Form the polynomials P and Q, note that these should be real
Q = numpy.poly(rQ);
P = numpy.poly(rP);
# Form the sum and difference filters by including known roots at z = 1 and
# z = -1
if p%2:
# Odd order: z = +1 and z = -1 are roots of the difference filter, P1(z)
P1 = numpy.convolve(P, [1, 0, -1])
Q1 = Q
else:
# Even order: z = -1 is a root of the sum filter, Q1(z) and z = 1 is a
# root of the difference filter, P1(z)
P1 = numpy.convolve(P, [1, -1])
Q1 = numpy.convolve(Q, [1, 1])
# Prediction polynomial is formed by averaging P1 and Q1
a = .5 * (P1+Q1)
return a[0:-1:1] # do not return last element
def find_coeff(poles, mode="f"):
"""
Find LP coefficients from a set of LP roots (poles).
Parameters
----------
poles : ndarray
Array of LP roots (poles)
mode : {'f', 'b'}
Mode in which LP coefficients should be returned. 'f' for coefficients
ordered m, m - 1,..., 1. 'b' for coefficients ordered 1, 2, ...., m.
Returns
-------
c : ndarray
LP coefficients ordered according to `mode`.
"""
# STABILITY
# the algorithm used here is numpy poly function which convolves
# the roots to find the coefficients, the accuracy of this method
# depends on the dtype of the poles parameter.
if mode not in ['f', 'b']:
raise ValueError("mode must be 'f'or 'b'")
if mode == 'f': # reverse resulting coefficients
return np.squeeze(-np.poly(poles)[:0:-1])
else: # keep coefficients as is
return np.squeeze(-np.poly(poles)[1:])
def compute_filter(self): theta = 2 * np.pi * self.frequency / self.nyquist * 2 zero = np.exp(np.array([1j, -1j]) * theta) pole = self.module_pole * zero self._filter_b = np.poly(zero) self._filter_a = np.poly(pole)
def lsf_to_lpc(all_lsf):
if len(all_lsf.shape) < 2:
all_lsf = all_lsf[None]
order = all_lsf.shape[1]
all_lpc = np.zeros((len(all_lsf), order + 1))
for i in range(len(all_lsf)):
lsf = all_lsf[i]
zeros = np.exp(1j * lsf)
sum_zeros = zeros[::2]
diff_zeros = zeros[1::2]
sum_zeros = np.hstack((sum_zeros, np.conj(sum_zeros)))
diff_zeros = np.hstack((diff_zeros, np.conj(diff_zeros)))
sum_filt = np.poly(sum_zeros)
diff_filt = np.poly(diff_zeros)
if order % 2 != 0:
deconv_diff = sg.convolve(diff_filt, [1, 0, -1])
deconv_sum = sum_filt
else:
deconv_diff = sg.convolve(diff_filt, [1, -1])
deconv_sum = sg.convolve(sum_filt, [1, 1])
lpc = .5 * (deconv_sum + deconv_diff)
# Last coefficient is 0 and not returned
all_lpc[i] = lpc[:-1]
return np.squeeze(all_lpc)
def do_plot(ax, z):
tt=np.linspace(0,1,100)
yy = z[0]*(1-tt)+z[1]*(tt)
z1=z[:5]
z2=z[5:]
P = np.poly(z1)
Q = np.poly(z2)
Z0 = np.roots((P-Q)[1:])
def Z(t):
F=P*(1-t)+Q*t
return np.roots(F)
N = 100
ZZ = np.zeros([5, N], np.complex)
for i in range(N):
ZZ[:,i]=Z(i/50.0-0.5).transpose()
rval = []
rval.extend( ax.plot(Z0.real, Z0.imag, "gx") )
rval.extend( ax.plot(ZZ.real.transpose(), ZZ.imag.transpose(), "r.") )
return rval
def apply_to(self, array, samplerate):
# Nyquist frequency
nyquist = samplerate / 2.
# ratio of notch freq. to Nyquist freq.
freq_ratio = self.frequency / nyquist
# width of the notch
notchWidth = 0.01
# Compute zeros
zeros = np.array([np.exp(1j * np.pi * freq_ratio),
np.exp(-1j * np.pi * freq_ratio)])
# Compute poles
poles = (1 - notchWidth) * zeros
b = np.poly(zeros) # Get moving average filter coefficients
a = np.poly(poles) # Get autoregressive filter coefficients
try:
if array.ndim == 1:
filtered = scipy.signal.filtfilt(b, a, array, padtype="even")
else:
filtered = np.empty(array.shape)
for i in range(array.shape[1]):
filtered[:, i] = scipy.signal.filtfilt(
b, a, array[:, i], padtype="even")
return filtered * self.gain + array * (1 - self.gain)
except ValueError:
print "Could not apply filter"
return array
def test_infnorm(): """Test function for infnorm()""" # FIXME M/P test needed num, den = np.poly([3, 0.3, 1]), np.poly([2, 0.5, .25]) H = (num, den) Hinf, fmax = infnorm(H) assert np.allclose((Hinf, fmax), (np.array([ 1.84888889]), np.array([ 0.50000001])), atol=1e-8, rtol=1e-5)
def test_impL1_num_den(self):
"""Test function for impL1() 2/4"""
sys1 = (np.array([-.4]), np.array([0, 0]), 1)
num = np.poly(sys1[0])
den = np.poly(sys1[1])
num[0] *= sys1[2]
tf = (num, den)
r3 = ds.impL1(tf, n=10)
self.assertTrue(np.allclose(self.r2, r3, atol=1e-8, rtol=1e-4))
def impz(self, zeros, poles, L):
from scipy.signal import lfilter
a = np.poly(poles)
b = np.poly(zeros)
d = np.zeros(L)
d[0] = 1
h = lfilter(b, a, d)
return h
def getTheta(ar_roots, ma_roots, sigma): '''Return theta for Libcarma''' ma_poly = sigma*np.poly(ma_roots) #TODO Change * to / ar_poly = np.poly(ar_roots) theta = np.concatenate((ar_poly[1:], ma_poly)) p = len(ar_roots) q = len(ma_roots) return theta, p, q
def ss2tf(A, B, C, D, input=0):
"""State-space to transfer function.
Parameters
----------
A, B, C, D : ndarray
State-space representation of linear system.
input : int, optional
For multiple-input systems, the input to use.
Returns
-------
num : 2-D ndarray
Numerator(s) of the resulting transfer function(s). `num` has one row
for each of the system's outputs. Each row is a sequence representation
of the numerator polynomial.
den : 1-D ndarray
Denominator of the resulting transfer function(s). `den` is a sequence
representation of the denominator polynomial.
"""
# transfer function is C (sI - A)**(-1) B + D
A, B, C, D = map(asarray, (A, B, C, D))
# Check consistency and make them all rank-2 arrays
A, B, C, D = abcd_normalize(A, B, C, D)
nout, nin = D.shape
if input >= nin:
raise ValueError("System does not have the input specified.")
# make MOSI from possibly MOMI system.
if B.shape[-1] != 0:
B = B[:, input]
B.shape = (B.shape[0], 1)
if D.shape[-1] != 0:
D = D[:, input]
try:
den = poly(A)
except ValueError:
den = 1
if (product(B.shape, axis=0) == 0) and (product(C.shape, axis=0) == 0):
num = numpy.ravel(D)
if (product(D.shape, axis=0) == 0) and (product(A.shape, axis=0) == 0):
den = []
return num, den
num_states = A.shape[0]
type_test = A[:, 0] + B[:, 0] + C[0, :] + D
num = numpy.zeros((nout, num_states + 1), type_test.dtype)
for k in range(nout):
Ck = atleast_2d(C[k, :])
num[k] = poly(A - dot(B, Ck)) + (D[k] - 1) * den
return num, den
def dual_band(F, P, E, eps, eps_R=1, x1=0.5, map_type=1):
r"""
Function to give modified F, P, and E polynomials after lowpass
prototype dual band transformation.
:param F: Polynomial F, i.e., numerator of S11 (in s-domain)
:param P: Polynomial P, i.e., numerator of S21 (in s-domain)
:param E: polynomial E is the denominator of S11 and S21 (in s-domain)
:param eps: Constant term associated with S21
:param eps_R: Constant term associated with S11
:param x1:
:param map_type:
:rtype:
"""
if (map_type == 1) or (map_type == 2):
s = sp.Symbol("s")
if map_type == 1:
a = -2j / (1 - x1 ** 2)
b = -1j * (1 + x1 ** 2) / (1 - x1 ** 2)
elif map_type == 2:
a = 2j / (1 - x1 ** 2)
b = 1j * (1 + x1 ** 2) / (1 - x1 ** 2)
s1 = a * s ** 2 + b
F = sp.Poly(F.ravel().tolist(), s)
F1 = sp.simplify(F.subs(s, s1))
F1 = sp.Poly(F1, s).all_coeffs()
F1 = I_to_i(F1)
E = sp.Poly(E.ravel().tolist(), s)
E1 = sp.simplify(E.subs(s, s1))
E1 = sp.Poly(E1, s).all_coeffs()
E1 = I_to_i(E1)
P = sp.Poly(P.ravel().tolist(), s)
P1 = sp.simplify(P.subs(s, s1))
P1 = sp.Poly(P1, s).all_coeffs()
P1 = I_to_i(P1)
elif map_type == 3:
F_roots = np.roots(F.ravel().tolist())
P_roots = np.roots(P.ravel().tolist())
F1_roots = Lee_roots_map(F_roots, x1)
P1_roots = Lee_roots_map(P_roots, x1)
P1_roots = np.concatenate((P1_roots, np.array([0, 0])))
F1 = np.poly(F1_roots)
P1 = np.poly(P1_roots)
F1 = np.reshape(F1, (len(F1), -1))
P1 = np.reshape(P1, (len(P1), -1))
E1 = poly_E(eps, eps_R, F1, P1)[0]
return F1, P1, E1
def solve(self):
if len(self.pole) == 0:
self.a = np.ones((1))
else:
self.a = np.poly(self.pole)
if len(self.zero) == 0:
self.b = np.ones((1))
else:
self.b = np.poly(self.zero)
self.solved = True
def init(self, freq=50.0, mod=0.9):
theta = 2 * np.pi * 50 / self.input.sample_rate
zero = np.exp(np.array([1j, -1j]) * theta)
pole = mod * zero
a, b = np.poly(pole), np.poly(zero)
#notch_ab = numpy.poly(zero), numpy.poly(pole)
#notch_ab = scipy.signal.iirfilter(32, [30.0 / 125], btype='low')
self.gr_block = filter.iir_filter_ffd(b, a, oldstyle=False)
def z_coeff(Poles,Zeros,fs,g,fg,fo = 'none'):
if fg == np.inf:
fg = fs/2
if fo == 'none':
beta = 1.0
else:
beta = f_warp(fo,fs)/fo
a = np.poly(z_from_f(beta*np.array(Poles),fs))
b = np.poly(z_from_f(beta*np.array(Zeros),fs))
gain = 10.**(g/20.)/abs(Fz_at_f(beta*np.array(Poles),beta*np.array(Zeros),fg,fs))
return (a,b*gain)
def invresz(r, p, k, tol=1e-3, rtype='avg'):
"""Compute b(z) and a(z) from partial fraction expansion: r,p,k
If M = len(b) and N = len(a)
b(z) b[0] + b[1] z**(-1) + ... + b[M-1] z**(-M+1)
H(z) = ------ = ----------------------------------------------
a(z) a[0] + a[1] z**(-1) + ... + a[N-1] z**(-N+1)
r[0] r[-1]
= --------------- + ... + ---------------- + k[0] + k[1]z**(-1) ...
(1-p[0]z**(-1)) (1-p[-1]z**(-1))
If there are any repeated roots (closer than tol), then the partial
fraction expansion has terms like
r[i] r[i+1] r[i+n-1]
-------------- + ------------------ + ... + ------------------
(1-p[i]z**(-1)) (1-p[i]z**(-1))**2 (1-p[i]z**(-1))**n
See also
--------
residuez, poly, polyval, unique_roots
"""
extra = asarray(k)
p, indx = cmplx_sort(p)
r = take(r, indx, 0)
pout, mult = unique_roots(p, tol=tol, rtype=rtype)
p = []
for k in range(len(pout)):
p.extend([pout[k]] * mult[k])
a = atleast_1d(poly(p))
if len(extra) > 0:
b = polymul(extra, a)
else:
b = [0]
indx = 0
brev = asarray(b)[::-1]
for k in range(len(pout)):
temp = []
# Construct polynomial which does not include any of this root
for l in range(len(pout)):
if l != k:
temp.extend([pout[l]] * mult[l])
for m in range(mult[k]):
t2 = temp[:]
t2.extend([pout[k]] * (mult[k] - m - 1))
brev = polyadd(brev, (r[indx] * poly(t2))[::-1])
indx += 1
b = real_if_close(brev[::-1])
return b, a
def invres(r,p,k,tol=1e-3,rtype='avg'):
"""Compute b(s) and a(s) from partial fraction expansion: r,p,k
If M = len(b) and N = len(a)
b(s) b[0] x**(M-1) + b[1] x**(M-2) + ... + b[M-1]
H(s) = ------ = ----------------------------------------------
a(s) a[0] x**(N-1) + a[1] x**(N-2) + ... + a[N-1]
r[0] r[1] r[-1]
= -------- + -------- + ... + --------- + k(s)
(s-p[0]) (s-p[1]) (s-p[-1])
If there are any repeated roots (closer than tol), then the partial
fraction expansion has terms like
r[i] r[i+1] r[i+n-1]
-------- + ----------- + ... + -----------
(s-p[i]) (s-p[i])**2 (s-p[i])**n
See Also
--------
residue, poly, polyval, unique_roots
"""
extra = k
p, indx = cmplx_sort(p)
r = take(r,indx,0)
pout, mult = unique_roots(p,tol=tol,rtype=rtype)
p = []
for k in range(len(pout)):
p.extend([pout[k]]*mult[k])
a = atleast_1d(poly(p))
if len(extra) > 0:
b = polymul(extra,a)
else:
b = [0]
indx = 0
for k in range(len(pout)):
temp = []
for l in range(len(pout)):
if l != k:
temp.extend([pout[l]]*mult[l])
for m in range(mult[k]):
t2 = temp[:]
t2.extend([pout[k]]*(mult[k]-m-1))
b = polyadd(b,r[indx]*poly(t2))
indx += 1
b = real_if_close(b)
while allclose(b[0], 0, rtol=1e-14) and (b.shape[-1] > 1):
b = b[1:]
return b, a
def zp_expr(s,Z,P):
pZ = np.poly(Z)
pP = np.poly(P)
a = np.polyval(pZ, s)
b = np.polyval(pP, s)
print "zp_expr"
print " a",a
print " b",b
return a / b
def test_objects(self):
from decimal import Decimal
p = np.poly1d([Decimal('4.0'), Decimal('3.0'), Decimal('2.0')])
p2 = p * Decimal('1.333333333333333')
assert_(p2[1] == Decimal("3.9999999999999990"))
p2 = p.deriv()
assert_(p2[1] == Decimal('8.0'))
p2 = p.integ()
assert_(p2[3] == Decimal("1.333333333333333333333333333"))
assert_(p2[2] == Decimal('1.5'))
assert_(np.issubdtype(p2.coeffs.dtype, np.object_))
p = np.poly([Decimal(1), Decimal(2)])
assert_equal(np.poly([Decimal(1), Decimal(2)]),
[1, Decimal(-3), Decimal(2)])
def zero(self):
"""Compute the zeros of a state space system."""
if self.inputs > 1 or self.outputs > 1:
raise NotImplementedError("StateSpace.zeros is currently \
implemented only for SISO systems.")
den = poly1d(poly(self.A))
# Compute the numerator based on zeros
#! TODO: This is currently limited to SISO systems
num = poly1d(poly(self.A - dot(self.B, self.C)) + ((self.D[0, 0] - 1) *
den))
return roots(num)
def build_TF(poles=[], zeros=[]):
"""Build a TF whose poles and zeros are given. Complex poles or zeros
are created by passing (wn,z) as a tuple in the pole or zero list."""
if np.isscalar(poles):
poles = [poles]
if np.isscalar(zeros):
zeros = [zeros]
all_poles = _unpack_complex(poles)
all_zeros = _unpack_complex(zeros)
num = np.poly(all_zeros)
den = np.poly(all_poles)
G = TF(num,den)
return G
def test_sacpaz_from_resp(self):
# The following two files were both extracted from a dataless
# seed file using rdseed
respfile = os.path.join(os.path.dirname(__file__),
'data', 'RESP.NZ.CRLZ.10.HHZ')
sacpzfile = os.path.join(os.path.dirname(__file__),
'data', 'SAC_PZs_NZ_CRLZ_HHZ')
# This is a rather lengthy test, in which the
# poles, zeros and the gain of each instrument response file
# are converted into the corresponding velocity frequency response
# function which have to be sufficiently close. Possibly due to
# different truncations in the RESP-formatted and SAC-formatted
# response files the frequency response functions are not identical.
tr1 = Trace()
tr2 = Trace()
attach_resp(tr1, respfile, torad=True, todisp=False)
attach_paz(tr2, sacpzfile, torad=False, tovel=True)
p1 = tr1.stats.paz.poles
z1 = tr1.stats.paz.zeros
g1 = tr1.stats.paz.gain
t_samp = 0.01
n = 32768
fy = 1 / (t_samp * 2.0)
# start at zero to get zero for offset/ DC of fft
f = np.arange(0, fy + fy / n, fy / n) # arange should includes fy
w = f * 2 * np.pi
s = 1j * w
a1 = np.poly(p1)
b1 = g1 * np.poly(z1)
h1 = np.polyval(b1, s) / np.polyval(a1, s)
h1 = np.conj(h1)
h1[-1] = h1[-1].real + 0.0j
p2 = tr2.stats.paz.poles
z2 = tr2.stats.paz.zeros
g2 = tr2.stats.paz.gain
a2 = np.poly(p2)
b2 = g2 * np.poly(z2)
h2 = np.polyval(b2, s) / np.polyval(a2, s)
h2 = np.conj(h2)
h2[-1] = h2[-1].real + 0.0j
amp1 = abs(h1)
amp2 = abs(h2)
phase1 = np.unwrap(np.arctan2(-h1.imag, h1.real))
phase2 = np.unwrap(np.arctan2(-h2.imag, h2.real))
np.testing.assert_almost_equal(phase1, phase2, decimal=4)
rms = np.sqrt(np.sum((amp1 - amp2) ** 2) /
np.sum(amp2 ** 2))
self.assertTrue(rms < 2.02e-06)
self.assertTrue(tr1.stats.paz.t_shift, 0.4022344)
def lsf2poly(L):
order = len(L)
Q = L[::2]
P = L[1::2]
poles_P = np.r_[np.exp(1j * P), np.exp(-1j * P)]
poles_Q = np.r_[np.exp(1j * Q), np.exp(-1j * Q)]
P = np.poly(poles_P)
Q = np.poly(poles_Q)
P = convolve(P, np.array([1.0, -1.0]))
Q = convolve(Q, np.array([1.0, 1.0]))
a = 0.5 * (P + Q)
return a[:-1]
def test_simple(self):
z_r = np.array([0.5, -0.5])
p_r = np.array([1.j / np.sqrt(2), -1.j / np.sqrt(2)])
# Sort the zeros/poles so that we don't fail the test if the order
# changes
z_r.sort()
p_r.sort()
b = np.poly(z_r)
a = np.poly(p_r)
z, p, k = tf2zpk(b, a)
z.sort()
p.sort()
assert_array_almost_equal(z, z_r)
assert_array_almost_equal(p, p_r)
def setup_main_screen(self):
import matplotlib.gridspec as gridspec
# Poles & zeros
self.fig = plt.figure()
gs = gridspec.GridSpec(3, 12)
# self.ax = self.fig.add_axes([0.1, 0.1, 0.77, 0.77], polar=True, axisbg='#d5de9c')
# self.ax=self.fig.add_subplot(221,polar=True, axisbg='#d5de9c')
self.ax = plt.subplot(gs[0:, 0:6], polar=True, axisbg="#d5de9c")
# self.ax = self.fig.add_subplot(111, polar=True)
self.fig.suptitle(
"Poles & zeros adjustment", fontsize=18, color="blue", x=0.1, y=0.98, horizontalalignment="left"
)
# self.ax.set_title('Poles & zeros adjustment',fontsize=16, color='blue')
self.ax.set_ylim([0, self.ymax])
self.poles_line, = self.ax.plot(self.poles_th, self.poles_r, "ob", ms=9, picker=5, label="Poles")
self.zeros_line, = self.ax.plot(self.zeros_th, self.zeros_r, "Dr", ms=9, picker=5, label="Zeros")
self.ax.plot(np.linspace(-np.pi, np.pi, 500), np.ones(500), "--b", lw=1)
self.ax.legend(loc=1)
# Transfer function
# self.figTF, self.axTF = plt.subplots(2, sharex=True)
# self.axTF0=self.fig.add_subplot(222,axisbg='LightYellow')
self.axTF0 = plt.subplot(gs[0, 6:11], axisbg="LightYellow")
# self.axTF[0].set_axis_bgcolor('LightYellow')
self.axTF0.set_title("Transfer function (modulus)")
# self.axTF1=self.fig.add_subplot(224,axisbg='LightYellow')
self.axTF1 = plt.subplot(gs[1, 6:11], axisbg="LightYellow")
self.axTF1.set_title("Transfer function (phase)")
self.axTF1.set_xlabel("Frequency")
f = np.linspace(0, 1, self.N)
self.TF = np.fft.fft(np.poly(self.zeros), self.N) / np.fft.fft(np.poly(self.poles), self.N)
self.TF_m_line, = self.axTF0.plot(f, np.abs(self.TF))
self.TF_p_line, = self.axTF1.plot(f, 180 / np.pi * np.angle(self.TF))
# self.figTF.canvas.draw()
# Impulse response
# self.figIR = plt.figure()
# self.axIR = self.fig.add_subplot(223,axisbg='Lavender')
self.axIR = plt.subplot(gs[2, 6:11], axisbg="Lavender")
self.IR = self.impz(self.zeros, self.poles, self.Nir) # np.real(np.fft.ifft(self.TF))
self.axIR.set_title("Impulse response")
self.axIR.set_xlabel("Time")
self.IR_m_line, = self.axIR.plot(self.IR)
# self.figIR.canvas.draw()
self.fig.canvas.draw()
self.fig.tight_layout()
def design_allpole(pole_mags, pole_freqs, fs):
poles = []
for n in range(len(pole_mags)):
w_p = pole_freqs[n] / fs * (2 * np.pi)
poles.append(pole_mags[n] * np.exp(1j * w_p))
a = np.poly(np.asarray(poles))
return np.array([1]), a
def reachable_form(xsys):
# Check to make sure we have a SISO system
if not issiso(xsys):
raise ControlNotImplemented(
"Canonical forms for MIMO systems not yet supported")
# Create a new system, starting with a copy of the old one
zsys = StateSpace(xsys)
# Generate the system matrices for the desired canonical form
zsys.B = zeros(shape(xsys.B))
zsys.B[0, 0] = 1
zsys.A = zeros(shape(xsys.A))
Apoly = poly(xsys.A) # characteristic polynomial
for i in range(0, xsys.states):
zsys.A[0, i] = -Apoly[i + 1] / Apoly[0]
if (i + 1 < xsys.states):
zsys.A[i + 1, i] = 1
# Compute the reachability matrices for each set of states
Wrx = ctrb(xsys.A, xsys.B)
Wrz = ctrb(zsys.A, zsys.B)
# Transformation from one form to another
Tzx = Wrz * inv(Wrx)
# Finally, compute the output matrix
zsys.C = xsys.C * inv(Tzx)
return zsys, Tzx
def function_transfere(aircraft, point_trim):
alt, Ma, sm, km = point_trim
aircraft.set_mass_and_static_margin(km, sm)
va = dyn.va_of_mach(Ma, alt)
xtrim, utrim = dyn.trim(aircraft, {'va': va, 'gamm': 0, 'h': alt})
A, B = utils.num_jacobian(xtrim, utrim, aircraft, dyn.dyn)
A_4, B_4 = A[2:,2:], B[2:,:2][:,0].reshape((4,1))
C_4 = np.zeros((1,4))
C_4[0][2] = 1
Acc_4 = np.zeros((4, 4))
Bcc_4 = np.zeros((4, 1))
Bcc_4[3,0] = 1
valeursp, M = np.linalg.eig(A_4)
coeff = np.poly(valeursp)
for i in range(3):
Acc_4[i,i+1] = 1
Acc_4[3,3-i] = -coeff[i+1]
Acc_4[3,0] = -coeff[4]
Qccc = np.zeros((4, 4))
Q = np.zeros((4, 4))
for i in range(4):
Qccc[:,i:i+1] = np.dot(np.linalg.matrix_power(Acc_4, i), Bcc_4)
Q[:,i:i+1] = np.dot(np.linalg.matrix_power(A_4, i), B_4)
Mcc = np.dot(Q, np.linalg.inv(Qccc))
Ccc_4 = np.dot(C_4, Mcc)
num = list(Ccc_4)
num.reverse()
den = list(-Acc_4[3, :])
den.append(1.)
den.reverse()
return num, den, valeursp
def sample_data(self, rate, AR = 1):
'''
Generate simulated data with default class params. Feel free to change if needed. Can be AR1 or AR2
'''
if np.shape(rate)[-1] != self.Tps:
raise ValueError("rate should be the same length as T in the calcium class, got {0}".format(np.shape(data)[-1]))
spikes = onp.random.poisson(rate)
roots = onp.exp(-self.dt/np.array(self.AR_params)) ### here self.AR_params is an n-length vector where n is the AR process order
coeffs = onp.poly(roots) #find coefficients with of a polynomial with roots of params
z = lfilter([self.alpha],coeffs,spikes) #generate AR-n process with inputs spk_counts, where self.alpha is the CA increase due to a spike
# q = [exp(-self.dt/tau_samp),exp(-self.dt/tau_samp_b)]
# a_true = onp.poly(q)
# z = filter(1,a_true,spcounts);
# ##### To Do: AR2 PROCESS HERE!
trace = z + onp.sqrt(self.Gauss_sigma)*onp.random.randn(onp.size(z))#add noise
return trace, spikes
def principal_invariants(self):
"""
Returns a list of principal invariants for the tensor,
which are the values of the coefficients of the characteristic
polynomial for the matrix
"""
return np.poly(self)[1:] * np.array([-1, 1, -1])
def reachable_form(xsys):
# Check to make sure we have a SISO system
if not control.issiso(xsys):
raise control.ControlNotImplemented(
"Canonical forms for MIMO systems not yet supported")
# Create a new system, starting with a copy of the old one
zsys = control.StateSpace(xsys)
# Generate the system matrices for the desired canonical form
zsys.B = zeros(shape(xsys.B)); zsys.B[0, 0] = 1;
zsys.A = zeros(shape(xsys.A))
Apoly = poly(xsys.A) # characteristic polynomial
for i in range(0, xsys.states):
zsys.A[0, i] = -Apoly[i+1] / Apoly[0]
if (i+1 < xsys.states): zsys.A[i+1, i] = 1
# Compute the reachability matrices for each set of states
Wrx = control.ctrb(xsys.A, xsys.B)
Wrz = control.ctrb(zsys.A, zsys.B)
# Transformation from one form to another
Tzx = Wrz * inv(Wrx)
# Finally, compute the output matrix
zsys.C = xsys.C * inv(Tzx)
return zsys, Tzx
def principal_invariants(self):
"""
Returns a list of principal invariants for the tensor,
which are the values of the coefficients of the characteristic
polynomial for the matrix
"""
return np.poly(self)[1:] * np.array([-1, 1, -1])
def Filter(N, fc, fs):
N = float(N)
#Nth order butterworth filter
fc = float(fc)
#cutoff frequency
fs = float(fs)
#sampling frequency
wc = float(2 * math.pi * fc / fs)
T = 1 / fs
omega_c = (2 / T) * math.tan(wc / 2)
omega_c = 2 * math.pi * fc
s_k = [
omega_c *
cmath.exp(cmath.sqrt(-1) * math.pi * (N + 2 * k - 1) / (2 * N))
for k in range(1,
int(N) + 1)
]
s_k_poly = np.poly(s_k)
[a_k, p, s] = scipy.signal.residue(math.pow(omega_c, N), s_k_poly)
return [
sum([
T * a_k[k] /
(1 - cmath.exp(p[k] * T) * cmath.exp(-1 * cmath.sqrt(-1) * W[w]))
for k in range(0, int(N))
]) for w in range(0, len(W))
]
def ARHarmonicPoly(f0, rate, mag=0.99):
"""
Generates coefficients of an AR polynomial corresponding to a
harmonic excitation at frequency f0
"""
omega = float(f0)/rate * 2.0 * np.pi
n = int(np.pi / omega)
roots = np.ndarray((n*2), dtype='complex')
for i in range(n):
a = omega * (i+1)
c = mag*np.cos(a)
s = mag*np.sin(a)
roots[i*2] = complex(c, s)
roots[i*2+1] = complex(c, -s)
if True:
poly = np.poly(roots)[1:]
else:
# polyfromroots() returns the trailing coeff = 1, so just
# knock off that one and reverse the rest as we have negative
# powers.
poly = pn.polyfromroots(roots).real[-2::-1]
if np.max(np.abs(poly)) >= 1.0:
print(roots, np.prod(roots), poly)
raise OverflowError('Poly too big')
return -poly
def transfunc(self, polos, **kwargs):
wp = kwargs.get('wp', 0)
w0 = kwargs.get('w0', 0)
Bw = kwargs.get('bw', self.Bp)
ordem = kwargs.get('ord', self.ordem())
resp = kwargs.get('response', self.tipo)
G_db = kwargs.get('G', self.G_bp)
fcn = 0
if resp == 'lp':
self.den_norm = np.real(np.poly(polos))
denm = np.zeros(len(self.den_norm))
for i in range(0, len(polos) + 1):
denm[i] = self.den_norm[i] * np.power(wp, i)
if (ordem % 2 == 0):
num = denm[-1] * (1 / np.sqrt(1 + np.power(self.eps, 2)))
else:
num = denm[-1]
fcn = signal.TransferFunction(pow(10, -G_db / 20.0) * num, denm)
if resp == 'hp':
self.den_norm = np.real(np.poly(polos))
denm = np.zeros(len(polos) + 1)
for i in range(0, len(polos) + 1):
denm[i] = self.den_norm[len(polos) - i] * np.power(wp, i)
num = np.zeros(len(polos) + 1)
if (self.ordem() % 2 == 0):
num[0] = denm[0] * (1 / np.sqrt(1 + np.power(self.eps, 2)))
else:
num[0] = denm[0]
fcn = signal.TransferFunction(pow(10, -G_db / 20.0) * num, denm)
if (resp == 'bp'):
self.den_norm = np.real(np.poly(polos))
[num, den] = signal.lp2bp(self.den_norm[-1], self.den_norm, w0, Bw)
if (self.ordem() % 2 == 0):
num[0] = num[0] * (1 / np.sqrt(1 + np.power(self.eps, 2)))
else:
num[0] = num[0]
fcn = signal.TransferFunction(num, den)
if (resp == 'bs'):
self.den_norm = np.real(np.poly(polos))
[num, den] = signal.lp2bs(self.den_norm[-1], self.den_norm, w0, Bw)
if (self.ordem() % 2 == 0):
num = num[::] * (1 / np.sqrt(1 + np.power(self.eps, 2)))
else:
num[-1] = num[-1]
fcn = signal.TransferFunction(num, den)
self.fcn = fcn
return fcn
def ss2tf(A, B, C, D, input=0):
"""State-space to transfer function.
Inputs:
A, B, C, D -- state-space representation of linear system.
input -- For multiple-input systems, the input to use.
Outputs:
num, den -- Numerator and denominator polynomials (as sequences)
respectively.
"""
# transfer function is C (sI - A)**(-1) B + D
A, B, C, D = map(asarray, (A, B, C, D))
# Check consistency and
# make them all rank-2 arrays
A, B, C, D = abcd_normalize(A, B, C, D)
nout, nin = D.shape
if input >= nin:
raise ValueError, "System does not have the input specified."
# make MOSI from possibly MOMI system.
if B.shape[-1] != 0:
B = B[:,input]
B.shape = (B.shape[0],1)
if D.shape[-1] != 0:
D = D[:,input]
den = poly(A)
if (product(B.shape,axis=0) == 0) and (product(C.shape,axis=0) == 0):
num = numpy.ravel(D)
if (product(D.shape,axis=0) == 0) and (product(A.shape,axis=0) == 0):
den = []
end
return num, den
num_states = A.shape[0]
type_test = A[:,0] + B[:,0] + C[0,:] + D
num = numpy.zeros((nout, num_states+1), type_test.dtype)
for k in range(nout):
Ck = atleast_2d(C[k,:])
num[k] = poly(A - dot(B,Ck)) + (D[k]-1)*den
return num, den
def estimate_time_constant(y, p=2, sn=None, lags=10, fudge_factor=1., nonlinear_fit=False):
"""
Estimate AR model parameters through the autocovariance function
Parameters
----------
y : array, shape (T,)
One dimensional array containing the fluorescence intensities with
one entry per time-bin.
p : positive integer
order of AR system
sn : float
sn standard deviation, estimated if not provided.
lags : positive integer
number of additional lags where he autocovariance is computed
fudge_factor : float (0< fudge_factor <= 1)
shrinkage factor to reduce bias
Returns
-------
g : estimated coefficients of the AR process
"""
if sn is None:
sn = GetSn(y)
lags += p
# xc = axcov(y, lags)[lags:]
y = y - y.mean()
xc = np.array([y[i:].dot(y[:-i if i else None]) for i in range(1 + lags)]) / len(y)
if nonlinear_fit and p <= 2:
xc[0] -= sn**2
g1 = xc[:-1].dot(xc[1:]) / xc[:-1].dot(xc[:-1])
if p == 1:
def func(x, a, g):
return a * g**x
popt, pcov = curve_fit(func, list(range(len(xc))), xc, (xc[0], g1)) #, bounds=(0, [3 * xc[0], 1]))
return popt[1:2] * fudge_factor
elif p == 2:
def func(x, a, d, r):
return a * (d**(x + 1) - r**(x + 1) / (1 - r**2) * (1 - d**2))
popt, pcov = curve_fit(func, list(range(len(xc))), xc, (xc[0], g1, .1))
d, r = popt[1:]
d *= fudge_factor
return np.array([d + r, -d * r])
xc = xc[:, np.newaxis]
A = scipy.linalg.toeplitz(xc[np.arange(lags)],
xc[np.arange(p)]) - sn**2 * np.eye(lags, p)
g = np.linalg.lstsq(A, xc[1:], rcond=None)[0]
gr = np.roots(np.concatenate([np.array([1]), -g.flatten()]))
gr = (gr + gr.conjugate()) / 2.
gr[gr > 1] = 0.95 + np.random.normal(0, 0.01, np.sum(gr > 1))
gr[gr < 0] = 0.15 + np.random.normal(0, 0.01, np.sum(gr < 0))
g = np.poly(fudge_factor * gr)
g = -g[1:]
return g.flatten()
def test_minimal_realization_Transfer():
G = Transfer(
[1., -8., 28., -58., 67., -30.],
poly([1, 2, 3., 2, 3., 4, 1 + (2 + 1e-6) * 1j, 1 - (2 + 1e-6) * 1j]))
H_f = minimal_realization(G)
assert_almost_equal(H_f.num, array([[1]]))
H_nf = minimal_realization(G, tol=1e-7)
assert_almost_equal(H_nf.num, array([[1., -7., 21., -37., 30.]]))
def main():
plotdistinct = []
plotest = []
for i in range (12,30):
Nsamples = 2**i
polys = []
for count in range (1, Nsamples):
Clinks = np.zeros( (4*n, 4*n) )
for i in range(0,n):
jlist = []
for k in range(0, n):
o = random.randrange(n + 1, 3*n, 1)
jlist.append(o)
for j in range(1,len(jlist)):
Clinks[i][jlist[j]] = 1
for i in range(3*n, 4*n):
klist = []
for k in range(0, n):
o = random.randrange(n + 1, 3*n, 1)
klist.append(o)
for j in range(1,len(jlist)):
Clinks[klist[j]][i] = 1
Csparse = Clinks
for i in range(0,n):
for j in range(3*n + 1, 4*n):
Npaths = 0
for k in range (n+1, 3*n):
Npaths = Npaths + Csparse[i][k]*Csparse[k][j]
if(Npaths > 0):
Clinks[i][j] = 1
Csparse = Clinks
SAC = Csparse + np.transpose(Csparse)
ASAC = Csparse - np.transpose(Csparse)
comm = np.subtract(np.matmul(SAC, ASAC).shape, np.matmul(ASAC, SAC).shape)
a = np.matmul(SAC, ASAC).shape
b = np.matmul(ASAC, SAC).shape
comm = np.subtract(a,b)
polys.append(np.concatenate(((np.poly(SAC)), (np.poly(ASAC)), (np.poly(comm))), axis=None))
counter = tally(polys)
distinct = len(counter) + 1
print(Nsamples, " ", distinct)
def my_K(omega, xi=0.8):
omega_d = omega * math.sqrt(1 - xi ** 2)
p1 = complex(-omega * xi, omega_d)
p2 = complex(-omega * xi, -omega_d)
p3 = -1.5 * omega
poles = [p1, p2, p3]
coefs = -np.poly(poles)[::-1]
return np.real(coefs[:-1])
def polybyrow(R):
""" Calculate the polynomial coefficients for rows of roots in R
"""
nr, nc = np.shape(R)
A = np.zeros((nr, nc+1))
for i in range(nr):
A[i,] = np.real(np.poly(R[i,]))
return A
def nu(z1, z0, b, beta, s):
nu0 = ((b - z0) ** (1 - beta) - (b - z1) ** (1 - beta)) / (1 - beta)
nu = [nu0]
for i in range(1, s + 1):
c = (-1) ** i * np.poly([b] * i )[::-1]
nu_i = - sum([c[j] * ((b - z1) ** (j + 1 - beta) - (b - z0) ** (j + 1 - beta)) / (j + 1 - beta) for j in range(0, i + 1)])
nu.append(nu_i)
return nu
def mu(z1, z0, a, alpha, s):
mu0 = ((z1 - a) ** (1 - alpha) - (z0 - a) ** (1 - alpha)) / (1 - alpha)
mu = [mu0]
for i in range(1, s + 1):
c = np.poly([-a] * i)[::-1]
mu_i = sum([c[j] * ((z1 - a) ** (j + 1 - alpha) - (z0 - a) ** (j + 1 - alpha)) / (j + 1 - alpha) for j in range(0, i + 1)])
mu.append(mu_i)
return mu
def lagrange(nodes):
result = Polynomial([0])
w = Polynomial(np.poly(nodes)[::-1])
deriv = w.deriv(1)
for i in range(len(nodes)):
result = pl.polyadd(result,
make_l_k(i, nodes, w, deriv) * f(nodes[i]))[0]
return result
def test_polystab(self):
# test case
x = [1, -2, -3]
v = np.roots(x)
vs = 0.5 * (np.sign(np.abs(v) - 1) + 1)
v = (1 - vs) * v + vs / np.conj(v)
b = x[0] * np.poly(v)
self.assertTrue(np.all(IIRDesign.polystab(self.a) == b))
def mls_polynomial_coefficients(rho, degree):
"""Determine the coefficients for a MLS polynomial smoother
Parameters
----------
rho : {float}
Spectral radius of the matrix in question
degree : {int}
Degree of polynomial coefficients to generate
Returns
-------
Tuple of arrays (coeffs,roots) containing the
coefficients for the (symmetric) polynomial smoother and
the roots of polynomial prolongation smoother.
The coefficients of the polynomial are in descending order
References
----------
.. [1] Parallel multigrid smoothing: polynomial versus Gauss--Seidel
M. F. Adams, M. Brezina, J. J. Hu, and R. S. Tuminaro
J. Comp. Phys., 188 (2003), pp. 593--610
Examples
--------
>>> from pyamg.relaxation.chebyshev import mls_polynomial_coefficients
>>> mls = mls_polynomial_coefficients(2.0, 2)
>>> print mls[0] # coefficients
[ 6.4 -48. 144. -220. 180. -75.8 14.5]
>>> print mls[1] # roots
[ 1.4472136 0.5527864]
"""
# std_roots = np.cos(np.pi * (np.arange(degree) + 0.5)/ degree)
# print std_roots
roots = rho / 2.0 * (1.0 -
np.cos(2 * np.pi *
(np.arange(degree, dtype='float64') + 1) /
(2.0 * degree + 1.0)))
# print roots
roots = 1.0 / roots
# S_coeffs = list(-np.poly(roots)[1:][::-1])
S = np.poly(roots)[::-1] # monomial coefficients of S error propagator
SSA_max = rho / (
(2.0 * degree + 1.0)**2) # upper bound spectral radius of S^2A
S_hat = np.polymul(S, S) # monomial coefficients of \hat{S} propagator
S_hat = np.hstack(((-1.0 / SSA_max) * S_hat, [1]))
# coeff for combined error propagator \hat{S}S
coeffs = np.polymul(S_hat, S)
coeffs = -coeffs[:-1] # coeff for smoother
return (coeffs, roots)
def ratspec_to_ss(A, B, controllable=True):
q = A(0) / B(0)
LA = A.c / (np.power(1j, np.arange(len(A.c) - 1, -1, -1)))
LB = B.c / (np.power(1j, np.arange(len(B.c) - 1, -1, -1)))
#only take real coefficient from denumerator
LA = np.poly1d(LA.real)
LB = np.poly1d(LB.real)
#strangely np.roots seems flipped between real and imag
ra = LA.roots[LA.roots < 0]
rb = LB.roots[LB.roots < 0]
GA = np.poly(ra)
GB = np.poly(rb)
GA = np.poly1d(GA.real)
GB = np.poly1d(GB.real)
GA.c /= GA.c[-1]
GB.c /= GB.c[-1]
GA.c /= GB.c[0]
GB.c /= GB.c[0]
F = np.zeros((len(GB.c) - 1, len(GB.c) - 1), dtype=np.float64)
L = np.zeros((len(GB.c) - 1, 1), dtype=np.float64)
H = np.zeros((1, len(GB.c) - 1), dtype=np.float64)
if controllable:
F[-1, :] = -GB.c[-1:0:-1]
F[:-1, 1:] = np.eye(len(GB.c) - 2)
L[-1, 0] = 1
H[0, :len(GA.c)] = GA.c[-1::-1]
else:
F[:, -1] = -GB.c[-1:0:-1]
F[1:, :-1] = np.eye(len(GB.c) - 2)
L[:len(GA.c), 0] = GA.c[-1::-1]
H[0, -1] = 1
Pinf = sla.solve_lyapunov(F, -q * L @ L.T)
return F, L, q, H, Pinf
def _matched(sys_: TransferFunction,
sample_time: numbers.Real) -> Tuple[np.ndarray, ...]:
poles = sys_.pole()
zeros = sys_.zero()
num = np.poly(np.exp(zeros * sample_time))
den = np.poly(np.exp(poles * sample_time))
num = np.atleast_1d(num)
den = np.atleast_1d(den)
root_number_delta = np.roots(den).shape[0] - np.roots(num).shape[0]
while root_number_delta > 0:
num = np.polymul(num, np.array([1, 1]))
root_number_delta -= 1
nump = np.poly1d(num)
denp = np.poly1d(den)
ds = np.polyval(sys_.num, 0) / np.polyval(sys_.den, 0)
d1z = nump(1) / denp(1)
dz_gain = ds / d1z
return num * dz_gain, den
def lagrange(x):
n = len(x)
L = np.zeros((n, n))
for i in range(n):
y = np.delete(x, i)
polinomio = np.poly(y)
denominador = np.polyval(polinomio, x[i])
L[i] = polinomio / denominador
return L
def polyscale(a):
# we need to ensure that the roots of our filter have unit norm before computing z
# this ensures that the ratio of the input and output spectrum is a constant function
for i in range(a.shape[1]):
a[:, i] = np.flipud(a[:, i])
roots = np.roots(a[:, i])
a[:, i] = np.poly(np.divide(roots, np.abs(roots)))
a[:, i] = np.flipud(a[:, i])
return a
def fix_poles(self, a):
poles = np.roots(a)
THR = 0.98
if not np.any(np.abs(poles) > THR):
return a
idxs = np.abs(poles) > THR
poles[idxs] = THR * poles[idxs] / np.abs(poles[idxs])
return np.poly(poles)
def randomPureReal(N=4, M=3):
#Generate a random transfer function of degree N
num = np.poly(2 * np.random.rand(N) - 1)
den = np.poly(2 * np.random.rand(M) - 1)
G = ctrl.tf(num, den, 1)
realFreq = pureReal(G)
sizes = [abs(ctrl.evalfr(G, z)) for z in realFreq]
realFreq = realFreq[np.argsort(sizes)]
sizes = 10 * (np.arange(len(realFreq)) + 1)
#Plot the unit circle and all the values where it is pure real
x = np.linspace(0, 2 * np.pi)
plt.plot(np.cos(x), np.sin(x))
plt.scatter(realFreq.real, realFreq.imag, s=sizes, c='r')
plt.axis('scaled')
plt.show()
def daubcqf(n, dtype='min'):
"""
Function computes the Daubechies' scaling and wavelet filters
(normalized to sqrt(2)).
Input:
n : Length of filter (must be even)
dtype : Optional parameter that distinguishes the minimum phase,
maximum phase and mid-phase solutions ('min', 'max', or
'mid'). If no argument is specified, the minimum phase
solution is used.
Output:
h_0 : Minimal phase Daubechies' scaling filter
h_1 : Minimal phase Daubechies' wavelet filter
Example:
n = 4
dtype = 'min'
h_0, h_1 = daubcqf(n, dtype)
Example Result:
h_0 = array([0.4830, 0.8365, 0.2241, -0.1294])
h_1 = array([0.1294, 0.2241, -0.8365, 0.4830])
Reference: \"Orthonormal Bases of Compactly Supported Wavelets\",
CPAM, Oct.89
"""
if (n % 2 != 0):
raise Exception("No Daubechies filter exists for ODD length")
k = n // 2
a = p = q = 1
h_0 = np.array([1, 1])
for j in range(1, k):
a = -a * 0.25 * (j + k - 1) / j
h_0 = np.hstack((0, h_0)) + np.hstack((h_0, 0))
p = np.hstack((0, -p)) + np.hstack((p, 0))
p = np.hstack((0, -p)) + np.hstack((p, 0))
q = np.hstack((0, q, 0)) + a * p
q = np.sort(np.roots(q))
qt = q[0:k - 1]
if (dtype == 'mid'):
if (k % 2 == 1):
qt = np.hstack((q[0:n - 2:4], q[1:n - 2:4]))
else:
qt = np.hstack((q[0], q[3:k - 1:4], q[4:k - 1:4], q[n - 4:k:-4],
q[n - 5:k:-4]))
h_0 = np.convolve(h_0, np.real(np.poly(qt)))
h_0 = np.sqrt(2) * h_0 / sum(h_0)
if (dtype == 'max'):
h_0 = np.flipud(h_0)
if (np.abs(sum(np.power(h_0, 2))) - 1 > 1e-4):
raise Exception("Numerically unstable for this value of n")
h_1 = np.copy(np.flipud(h_0))
h_1[0:n - 1:2] = -h_1[0:n - 1:2]
return h_0, h_1
def chebpoly(n, x=None, kind=1):
"""
Return Chebyshev polynomial of the first or second kind.
These polynomials are orthogonal on the interval [-1,1], with
respect to the weight function w(x) = (1-x^2)^(-1/2+kind-1).
chebpoly(n) returns the coefficients of the Chebychev polynomial of degree N.
chebpoly(n,x) returns the Chebychev polynomial of degree N evaluated in X.
Parameters
----------
n : integer, scalar
degree of Chebychev polynomial.
x : array-like, optional
evaluation points
kind: 1 or 2, optional
kind of Chebychev polynomial (default 1)
Returns
-------
p : ndarray
polynomial coefficients if x is None.
Chebyshev polynomial evaluated at x otherwise
Examples
--------
>>> import numpy as np
>>> x = chebroot(3)
>>> np.abs(chebpoly(3,x))<1e-15
array([ True, True, True], dtype=bool)
>>> chebpoly(3)
array([ 4., 0., -3., 0.])
>>> x2 = chebroot(4,kind=2)
>>> np.abs(chebpoly(4,x2,kind=2))<1e-15
array([ True, True, True, True], dtype=bool)
>>> chebpoly(4,kind=2)
array([ 16., 0., -12., 0., 1.])
Reference
---------
http://en.wikipedia.org/wiki/Chebyshev_polynomials
"""
if x is None: # Calculate coefficients.
if n == 0:
p = np.ones(1)
else:
p = np.round(
pow(2, n - 2 + kind) * np.poly(chebroot(n, kind=kind)))
p[1::2] = 0
return p
else: # Evaluate polynomial in chebychev form
ck = np.zeros(n + 1)
ck[n] = 1.
return _chebval(np.atleast_1d(x), ck, kind=kind)
def botheigs(num, eigs, init):
charpoly = np.poly(eigs)
temp = []
for i in range(len(charpoly) - 1):
temp.append(-charpoly[i + 1])
print('\nThe coefficients are:\n')
print(temp)
a = bothways(num, init, temp)
return a
def power_spectrum(self, fs, p):
p = self.to_params(p)
ar_roots = self.ar_roots(p)
ma_roots = self.ma_roots(p)
ar_coefs = np.real(np.poly(ar_roots))
ma_coefs = np.real(np.poly(ma_roots))
ma_coefs /= ma_coefs[-1]
ma_coefs = ma_coefs[::-1]
s = par.bounded_values(p['logit_sigma'],
low=self.sigma_min,
high=self.sigma_max)
sigma = s / np.sqrt(cm.carma_variance(1.0, ar_roots, ma_coefs))
return cm.power_spectrum(fs, sigma, ar_coefs, ma_coefs)
def root_error(roots):
"""Make a scatter plot of the real and imaginary parts of
the real and imaginary parts of roots and of the real and
imaginary parts of the computed roots of the polynomial with
coefficients calculated so that it should have roots at
the same points as the original set of roots. """
computed_roots = np.poly1d(np.poly(roots)).roots
plt.scatter(roots.real, roots.imag, color='b')
plt.scatter(computed_roots.real, computed_roots.imag, color='r')
plt.show()
def setUp(self):
num = np.poly([3, 0.3, 1])
den = np.poly([2, 0.5, .25])
H = (num, den)
tstr1 = empty()
(tstr1.form, tstr1.num, tstr1.den) = ('coeff', num, den)
tstr2 = empty()
tstr2.form = 'zp'
(tstr2.zeros, tstr2.poles, tstr2.gain) = tf2zpk(num, den)
z = np.exp(1j * np.linspace(0, 2*np.pi, num=129, endpoint=True))
self.h1 = ds.evalTF(tstr1, z)
self.h2 = ds.evalTF(tstr2, z)
self.h3 = ds.evalTF(H, z)
self.h4 = ds.evalTF(lti(tstr2.zeros, tstr2.poles, tstr2.gain), z)
h5tf = lti(tstr2.zeros, tstr2.poles, tstr2.gain).to_ss()
self.h5 = ds.evalTF((h5tf.A, h5tf.B, h5tf.C, h5tf.D), z)
h6tf = np.vstack((np.hstack((h5tf.A, h5tf.B)),
np.hstack((h5tf.C, np.atleast_2d(h5tf.D)))))
self.h6 = ds.evalTF(h6tf, z)