def loadConfig(self):
"""
Load the Soapy config file
"""
self.config = confParse.loadSoapyConfig(self.configfile)
self.nIters = self.config.sim.nIters
self.mask = circle.circle(
self.config.sim.pupilSize/2., self.config.sim.simSize)
self.wfsPixelScale = self.config.wfss[0].subapFOV/self.config.wfss[0].pxlsPerSubap
# Initialise the WFS (can use same one for all tests)
self.wfs = wfs.ShackHartmann(self.config, mask=self.mask)
def loadConfig(self):
"""
Load the Soapy config file
"""
self.config = confParse.loadSoapyConfig(self.configfile)
self.nIters = self.config.sim.nIters
self.mask = circle.circle(self.config.sim.pupilSize / 2.,
self.config.sim.simSize)
self.wfsPixelScale = self.config.wfss[0].subapFOV / self.config.wfss[
0].pxlsPerSubap
# Initialise the WFS (can use same one for all tests)
self.wfs = wfs.Gradient(self.config, mask=self.mask)
gsPositions, layerHeights, cn2,
L0)
residual = theoCovMat - rawCovMat
print("\n***\nRMS: {}\n***\n\n".format(numpy.sqrt(
(residual**2).mean())))
if callback:
callback(theoCovMat)
return (residual**2).flatten()
if __name__ == "__main__":
mask = circle.circle(3.5, 7)
gsPositions = numpy.array([[1., 0], [0, 0], [-1., 0]])
gsAltitudes = numpy.array([0, 0, 0])
nLayers = 1
layerHeights = numpy.array([12376.])
cn2 = numpy.array([10.])
L0 = numpy.array([100.])
T = TomoAO(3, mask, 0.6)
covMat = T.makeCovMat(gsAltitudes, gsPositions, layerHeights, cn2, L0)
# T.fitToData(
# covMat, 1, gsAltitudes, gsPositions, layerHeights, cn2, L0,
# fitCn2=True)
TEL_DIAM = 4.2
NWFS = 4
NSUBAPS = numpy.array([749] * NWFS)
NXSUBAPS = numpy.array([31] * NWFS)
SUBAPDIAM = numpy.array([4.2 / NXSUBAPS[0]] * NWFS)
GSALT = numpy.array([0] * NWFS)
GSPOS = numpy.array([[0, 0], [0, 0.1], [0, -.1], [.1, 0]
]) * (1. / 3600) * (numpy.pi / 180.)
NLAYERS = 1
LAYERHEIGHTS = numpy.array([12376.])
CN2 = numpy.array([10.0])
L0 = numpy.array([10.])
PUPIL_MASK = circle.circle(15.5, 31)
# class LATomo(object):
# def __init__(self, pxlScale, totSubaps, nxSubaps, subapDiam):
# self.pxlScale = pxlScale
# self.totSubaps = totSubaps
# self.nxSubaps = nxSubaps
# self.subapMask = circle.circle(nxSubaps/2., nxSubaps)
# self.subapDiam = subapDiam
# def loadRawData(self, filename):
# self.rawSlopes = fits.getdata(filename)
# self.rawSlopes_rad = (self.rawSlopes *
# self.pxlScale * numpy.pi/180./3600.)
self.covMat = numpy.frombuffer(self.ffiTomo.buffer(c_covMat,
size=covMatSize**2),
dtype="float64")
return self.covMat
def cov_XX(du, dv, ac, ad, bc, bd, L0):
pass
if __name__ == "__main__":
covMat = MatCov("matcov.dylib")
xCoords, yCoords = numpy.where(circle.circle(3.5, 7))
xCoords = numpy.tile(xCoords, 2) * 4.2
yCoords = numpy.tile(yCoords, 2) * 4.2
nSubaps = numpy.array([37, 37])
nxSubaps = numpy.array([7, 7])
gsAlt = numpy.array([0, 0])
wfsType = numpy.array([1, 1])
gsPosX = numpy.array([0, 5])
gsPosY = numpy.array([0, -5])
subapDiam = numpy.array([0.6, 0.6])
def test_circle(self):
log.info("Create circles with (0,5), (1,5), (2,5), (0,6),...,(3,6)")
# Define the expected outputs:
# >>> circle.circle(0, 5)
ce = numpy.array([[0., 0., 0., 0., 0.], [0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0.], [0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0.]])
c = circle.circle(0, 5)
self.assertEqual(c.tolist(), ce.tolist())
self.assertTrue((c == ce).all())
# circle.circle(1, 5)
ce = numpy.array([[0., 0., 0., 0., 0.], [0., 0., 1., 0., 0.],
[0., 1., 1., 1., 0.], [0., 0., 1., 0., 0.],
[0., 0., 0., 0., 0.]])
c = circle.circle(1, 5)
self.assertEqual(c.tolist(), ce.tolist())
self.assertTrue((c == ce).all())
# circle.circle(2, 5)
ce = numpy.array([[0., 0., 1., 0., 0.], [0., 1., 1., 1., 0.],
[1., 1., 1., 1., 1.], [0., 1., 1., 1., 0.],
[0., 0., 1., 0., 0.]])
c = circle.circle(2, 5)
self.assertEqual(c.tolist(), ce.tolist())
self.assertTrue((c == ce).all())
# circle.circle(0, 6)
ce = numpy.array([[0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0.]])
c = circle.circle(0, 6)
self.assertEqual(c.tolist(), ce.tolist())
self.assertTrue((c == ce).all())
# circle.circle(1, 6)
ce = numpy.array([[0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0.],
[0., 0., 1., 1., 0., 0.], [0., 0., 1., 1., 0., 0.],
[0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0.]])
c = circle.circle(1, 6)
self.assertEqual(c.tolist(), ce.tolist())
self.assertTrue((c == ce).all())
# circle.circle(2, 6)
ce = numpy.array([[0., 0., 0., 0., 0., 0.], [0., 0., 1., 1., 0., 0.],
[0., 1., 1., 1., 1., 0.], [0., 1., 1., 1., 1., 0.],
[0., 0., 1., 1., 0., 0.], [0., 0., 0., 0., 0., 0.]])
c = circle.circle(2, 6)
self.assertEqual(c.tolist(), ce.tolist())
self.assertTrue((c == ce).all())
# circle.circle(3, 6)
ce = numpy.array([[0., 1., 1., 1., 1., 0.], [1., 1., 1., 1., 1., 1.],
[1., 1., 1., 1., 1., 1.], [1., 1., 1., 1., 1., 1.],
[1., 1., 1., 1., 1., 1.], [0., 1., 1., 1., 1., 0.]])
c = circle.circle(3, 6)
self.assertEqual(c.tolist(), ce.tolist())
self.assertTrue((c == ce).all())
# This is a crucial test since this is used in the dragon config file:
# circle(15.5, 31) - circle(3.7, 31)
ce = numpy.array([[
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0
],
[
0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0
],
[
0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0
],
[
0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0
],
[
0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0
],
[
0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0
],
[
0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0
],
[
0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0
],
[
0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0
],
[
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0
],
[
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0
],
[
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0
],
[
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0,
0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1
],
[
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0,
0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1
],
[
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0,
0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1
],
[
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0,
0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1
],
[
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0,
0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1
],
[
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0,
0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1
],
[
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0,
0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1
],
[
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0
],
[
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0
],
[
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0
],
[
0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0
],
[
0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0
],
[
0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0
],
[
0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0
],
[
0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0
],
[
0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0
],
[
0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0
],
[
0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0
],
[
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1,
1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
]])
c = circle.circle(15.5, 31) - circle.circle(3.7, 31)
self.assertEqual(c.tolist(), ce.tolist())
self.assertTrue((c == ce).all())
log.info("Create circles with (2,6,(0.5,0.5)), (1,5,(0.5,0.5)), etc.")
c1 = circle.circle(2, 6, (0.5, 0.5))[1:, 1:]
c2 = circle.circle(2, 5)
self.assertEqual(c1.tolist(), c2.tolist())
self.assertTrue((c1 == c2).all())
c1 = circle.circle(1, 6, (0.5, 0.5))[1:, 1:]
c2 = circle.circle(1, 5)
self.assertEqual(c1.tolist(), c2.tolist())
self.assertTrue((c1 == c2).all())
c1 = circle.circle(0, 6, (0.5, 0.5))[1:, 1:]
c2 = circle.circle(0, 5)
self.assertEqual(c1.tolist(), c2.tolist())
self.assertTrue((c1 == c2).all())
c1 = circle.circle(0, 5, (0.5, 0.5))[1:, 1:]
c2 = circle.circle(0, 6)[1:-1, 1:-1]
self.assertEqual(c1.tolist(), c2.tolist())
self.assertTrue((c1 == c2).all())
c1 = circle.circle(1, 5, (0.5, 0.5))[1:, 1:]
c2 = circle.circle(1, 6)[1:-1, 1:-1]
#c2[3,2] = 3.0 # to test the testing
self.assertEqual(c1.tolist(), c2.tolist())
self.assertTrue((c1 == c2).all())
c1 = circle.circle(2, 5, (0.5, 0.5))[1:, 1:]
c2 = circle.circle(2, 6)[1:-1, 1:-1]
self.assertEqual(c1.tolist(), c2.tolist())
self.assertTrue((c1 == c2).all())
log.info("Raise TypeError if inputs not of correct type")
# For testing by hand and observing the result:
if False:
import pylab
pylab.ion()
c = circle.circle(2, 5, (0.5, 0.5))
pylab.imshow(c, interpolation="nearest")
c1 = circle.circle(1, 5)
def makeCovMat_LGSUplink(subaps,
d,
r0,
h,
H,
mask,
samp=1000,
pos1=(0, 0),
pos2=(0, 0),
L0=None,
l0=None):
"""
Parameters:
subaps (int): No. of subaps in 1d on WFS.
d (float): diameter of each subap.
r0 (float): Fried param of layer.
h (float): height of layer.
H (float): height of Laser.
mask (ndarray): 2d array of subaps, 1 if active, 0 if not.
samp (int, opt): size of oversampled covariance map
pos1 (tuple, opt): Angular Pos of LGS 1, arcsecs.
pos2 (tuple, opt): Angular Pos of LGS 2, arcsecs.
L0 (float, opt): Outer scale.
l0 (float, opt): innterscale.
Returns:
ndarray: The covariance Matrix
"""
#Size of the telescope
D = d * subaps
h = float(h)
H = float(H)
pos1 = numpy.array(pos1, dtype="float32") * (1. / 3600) * (numpy.pi / 180.)
pos2 = numpy.array(pos2, dtype="float32") * (1. / 3600) * (numpy.pi / 180.)
#Find the maximum seperation in metres
maxSep = (pos2 - pos1).max() * h
maxSep += d * (subaps + 1)
#Make the massively oversampled covaranace maps - make it an odd number
#if samp%2==0:
# samp+=1
covMaps = makeCovMap(samp, maxSep, r0, subapDiam, L0, l0)
covMaps /= covMaps[0].max()
#The subap size at this height
dSize = d * (1. - (h / H))
#And set to same scale as covMap array
dSize *= (covMaps.shape[1] - 1) / (2 * maxSep)
dSize = int(round(dSize))
print(dSize)
#Find total number of subaps, and their positions in subap space.
#Make a placeholder of the delicious covariance matrix
totalSubaps = int(mask.sum())
covMat = numpy.zeros((2 * totalSubaps, 2 * totalSubaps))
#This is the start of the sub-ap seen by the LGS on the uplink, in metres
lgs2Pos = (h * pos2) - d / 2. #Would need to be changed for launch
#Make an LGS mask to determine the overlap with subaps, each element is a cm
lgsScale = 100
lgsMask = circle.circle(lgsScale * d / 2.,
lgsScale * d * subaps,
centre_offset=lgsScale * lgs2Pos)
lgsSubapPos = []
subapPos = []
for x in range(subaps):
for y in range(subaps):
if mask[x, y] == 1:
subapPos.append([x, y])
subapPos = numpy.array(subapPos)
lgsSubapPos = numpy.array(lgsSubapPos)
for i in range(totalSubaps):
for j in range(totalSubaps):
#This is the position in metres from the centre of the WFS field, i
#to the start of the subap
maskVal = lgsMask[subapPos[j, 0] * 100:(subapPos[j, 0] + 1) * 100,
subapPos[j, 1] * 100:(subapPos[j, 1] + 1) *
100].mean()
if maskVal > 0:
ix, iy = pos1 * h + ((subapPos[i] * d) - D / 2.) * (1. -
(h / H))
jx, jy = pos2 * h + ((subapPos[j] * d) - D / 2.) * (1. -
(h / H))
#The seperation between the two subap start positions in covMap pxls
dx = (jx - ix) * (covMaps.shape[1] - 1) / (2 * maxSep)
dy = (jy - iy) * (covMaps.shape[1] - 1) / (2 * maxSep)
#print("ix:{}, jx:{}, dx:{}, iy:{}, jy:{}, dy:{}".format(ix, jx, dx, iy, jy, dy))
#Adjust this for the array coords
dx = int(round(dx + covMaps.shape[1] / 2. - dSize / 2.))
dy = int(round(dy + covMaps.shape[1] / 2. - dSize / 2.))
#print("i:{}, j:{}, ix:{}, jx:{}, dx:{}, iy:{}, jy:{}, dy:{}\n".format(i,j,ix, jx, dx, iy, jy, dy))
#Get the sum of the covarance for the seperation of
#each point in the sub-apertures
#x-x terms
covMat[i, :totalSubaps] += covMaps[0, dy:dy + dSize, dx:dx +
dSize].sum() * maskVal
#y-y terms
covMat[i + totalSubaps,
totalSubaps:] += covMaps[1, dy:dy + dSize,
dx:dx + dSize].sum() * maskVal
#x-y terms
covMat[i, :totalSubaps] += covMaps[2, dy:dy + dSize, dx:dx +
dSize].sum() * maskVal
covMat[i + totalSubaps,
totalSubaps:] += covMaps[2, dx:dx + dSize,
dy:dy + dSize].sum() * maskVal
#return covMaps
return covMat
# Parameters for CANARY
TEL_DIAM = 4.2
NWFS = 2
NSUBAPS = numpy.array([37] * NWFS)
NXSUBAPS = numpy.array([7] * NWFS)
SUBAPDIAM = numpy.array([0.6] * NWFS)
GSALT = numpy.array([0] * NWFS)
GSPOS = numpy.array([[10., 0], [-10, 0]]) * (1. / 3600) * (numpy.pi / 180.)
NLAYERS = 1
LAYERHEIGHTS = numpy.array([12376.])
CN2 = numpy.array([10.0])
L0 = numpy.array([100.])
PUPIL_MASK = circle.circle(3.5, 7)
def canaryCovMat():
subapPos = (numpy.array(numpy.where(PUPIL_MASK == 1)).T * TEL_DIAM /
NXSUBAPS[0]).T
# subapPos = numpy.array(numpy.where(PUPIL_MASK==1))/float(NXSUBAPS[0])
subapPos = numpy.tile(subapPos, (1, NWFS))
print(subapPos)
covMat = numpy.zeros((2 * NSUBAPS.sum(), 2 * NSUBAPS.sum()),
dtype="float64")
covMat = fillCovMat(covMat, subapPos)
return covMat