import astropy.units as u
import numpy as np
from scipy.interpolate import interp1d
import exosim.models.signal as signal
import exosim.output as output
from exosim.tasks.task import Task
from exosim.utils import RunConfig, check_units
[docs]
class DarkCurrentMap(Task):
"""
Produces the channel dark current map
Returns
--------
:class:`~exosim.models.signal.Signal`
channel dark current map
Raises
------
TypeError:
if the output is not a :class:`~exosim.models.signal.Signal` class
Notes
-----
This is a default class with standardized inputs and outputs.
The user can load this class and overwrite the "model" method
to implement a custom Task to replace this.
"""
def __init__(self):
"""
Parameters
__________
parameters: dict
dictionary containing the parameters. This is usually parsed from :class:`~exosim.tasks.load.load_options.LoadOptions`
time: :class:`~astropy.units.Quantity`
time grid
output: :class:`~exosim.output.output.Output` (optional)
output file
"""
self.add_task_param("parameters", "parameters dict")
self.add_task_param("time", "time grid")
self.add_task_param("output", "output file", None)
[docs]
def execute(self):
self.info("creating dark current map")
parameters = self.get_task_param("parameters")
tt = self.get_task_param("time")
dc_map = self.model(parameters, tt)
# checking output
if not isinstance(dc_map, signal.Signal):
self.error("wrong output format")
raise TypeError("wrong output format")
self.debug(f"dark current map: {dc_map.data}")
output_file = self.get_task_param("output")
if output_file and issubclass(output_file.__class__, output.Output):
dc_map.write(output_file, "dc_map")
self.set_output(dc_map)
[docs]
def model(self, parameters, time):
"""
Parameters
----------
parameters: dict
dictionary contained the sources parameters. This is usually parsed from :class:`~exosim.tasks.load.load_options.LoadOptions`
time: :class:`~astropy.units.Quantity`
time grid.
Returns
--------
:class:`~exosim.models.signal.Signal`
dark current efficiency map
"""
# variance map
if "dc_sigma" not in parameters["detector"]:
self.error("dc_sigma not provided in detector configuration")
raise KeyError("dc_sigma not provided in detector configuration")
if "dc_mean" in parameters["detector"]:
pass
elif "dc_median" in parameters["detector"]:
check_units(parameters["detector"]["dc_median"], "ct/s")
self.compute_dc_mean(parameters["detector"])
check_units(parameters["detector"]["dc_mean"], "ct/s")
check_units(parameters["detector"]["dc_sigma"], "ct/s")
variance_map = RunConfig.random_generator.normal(
parameters["detector"]["dc_mean"].value,
parameters["detector"]["dc_sigma"].value,
(
1,
parameters["detector"]["spatial_pix"],
parameters["detector"]["spectral_pix"],
),
)
idx = np.where(variance_map <= 0)
variance_map[idx] = 1e-16
t = [0] * u.hr
# temporal variation
if "dc_aging_factor" in parameters["detector"]:
t0 = parameters["detector"]["dc_aging_time_scale"]
t0 = check_units(t0, u.hr, force=True)
# creates the aging effect
dc_aging = RunConfig.random_generator.normal(
1,
parameters["detector"]["dc_aging_factor"],
(
1,
parameters["detector"]["spatial_pix"],
parameters["detector"]["spectral_pix"],
),
)
# remove values greater than 1, because QE should become smaller
dc_aging[dc_aging > 1] = 2 - dc_aging[dc_aging > 1]
# create the map evolution over time
variance_map = np.vstack((variance_map, variance_map * dc_aging))
t = np.hstack((t, t0))
variance_map = variance_map.reshape(
(t.size, variance_map.shape[1] * variance_map.shape[2])
)
f = interp1d(t, variance_map, fill_value="extrapolate", axis=0)
variance_map = f(time)
variance_map = variance_map.reshape(
(
time.size,
parameters["detector"]["spatial_pix"],
parameters["detector"]["spectral_pix"],
)
)
t = time
# resized map considering an oversampling factor
resized_map = variance_map.repeat(
parameters["detector"]["oversampling"], axis=1
).repeat(parameters["detector"]["oversampling"], axis=2)
dc_map = signal.Dimensionless(
data=resized_map,
spectral=np.arange(0, parameters["detector"]["spectral_pix"]) * u.pix,
time=t,
)
dc_map.temporal_rebin(time)
return dc_map
[docs]
def compute_dc_mean(self, detector):
r"""
Computes the mean of the dark current (dc_mean) from the log-normal distribution.
Notes
-----
The probability density function for the log-normal distribution is:
.. math::
pdf(x) = \frac{1}{\sigma x \sqrt{2\pi}}
\exp\left(-\frac{(\log(x) - \mu)^2}{2 \sigma^2}\right)
The mean of the pdf can be computed as:
.. math::
mean = \exp\left(\mu + \frac{s^2}{2}\right)
where s is the pdf standard deviation, computed by taking the sqrt of the variance, defined as:
.. math::
var = \left(\exp(\sigma^2) - 1\right) \exp(2 \mu + \sigma^2)
Parameters
----------
detector: dict
Dictionary for the detector. This is usually parsed from :class:`~exosim.tasks.load.load_options.LoadOptions`
Returns
-------
None
Updates the detector dictionary with the value of dc_mean
"""
mu = np.log(detector["dc_median"].value)
def root(s):
return detector["dc_sigma"].value ** 2 - (np.exp(s**2) - 1) * np.exp(
2 * mu + s**2
)
from scipy.optimize import newton
sigma = newton(func=root, x0=0.5)
atol = 1e-6
if not np.isclose(root(sigma), 0.0, atol=atol):
self.error(f"dc_sigma root not close to 0 (tol={atol:.1e})")
raise ValueError(f"dc_sigma root not close to 0 (tol={atol:.1e})")
dc_mean = np.exp(mu + sigma**2 / 2)
detector.update({"dc_mean": dc_mean * u.ct / u.s})
