CorrelatedSignal

class sofia_redux.scan.signal.correlated_signal.CorrelatedSignal(integration, mode)

Bases: Signal

Initialize a correlated signal.

The correlated signal is used to extract a signal from the integration defined by the supplied correlated mode. The mode defines a set of instrument channels, and how gain should be determined from them.

Parameters:

integrationsofia_redux.scan.integration.integration.Integration

The integration to which this signal belongs.

modesofia_redux.scan.channels.mode.correlated_mode.CorrelatedMode

The correlated modality mode for this signal.

Methods Summary

calc_filtering()	Update the source filtering of the signal.
copy()	Return a copy of the signal.
differentiate()	Differentiate the signal values and weights in-place.
get_differential()	Return a differentiated copy of the signal.
get_frame_data_signal()	Get the signal as it would appear translated to the frame data.
get_integral()	Return an integrated copy of the signal.
get_mean()	Return the mean signal value and weight.
get_median()	Return the median signal value and weight.
get_ml_correlated(channel_group[, ...])	Get the maximum-likelihood correlated gain increment and weight.
get_parms()	Return the degrees of freedom for the signal.
get_robust_correlated(channel_group[, ...])	Get the median derived correlated gain increment and weight.
get_variance()	Return the signal variance.
integrate()	Integrate the signal values and weights in-place.
level([start_frame, end_frame, robust])	Remove the mean value from the signal values.
update([robust])	Update the frame data by the signal.
weight_at(frame)	Return the weight at a given frame index.

Methods Documentation

calc_filtering()

Update the source filtering of the signal.

Where phi is the channel dependents, they are updated by:

phi = mean(phi + phi * channel_overlaps)

The signal source filtering is set to 1 - phi. The channel source filtering has the prior correction undone, and then updated as:

csf *= signal source filtering.

Returns:

None

copy()

Return a copy of the signal.

Returns:

CorrelatedSignal

differentiate()

Differentiate the signal values and weights in-place.

Note that the spacing between sample values is assumed to be in seconds when calculating the gradient.

Returns:

None

get_differential()

Return a differentiated copy of the signal.

Returns:

CorrelatedSignal

get_frame_data_signal()

Get the signal as it would appear translated to the frame data.

Returns:

frame_data_signalnumpy.ndarray (float)

The signal contribution to the frame data of shape (n_frames, n_channels).

get_integral()

Return an integrated copy of the signal.

Returns:

CorrelatedSignal

get_mean()

Return the mean signal value and weight.

Returns:

mean, mean_weight

get_median()

Return the median signal value and weight.

Returns:

median, median_weightfloat, float

get_ml_correlated(channel_group, modeling_frames=None)

Get the maximum-likelihood correlated gain increment and weight.

The gain increments are given as:

dC_s = sum_{f | s}{w_f * sum_{c} {w_c g_c d_{f,c}} / dW_s

where dW_s is the gain increment weight given by

dW_s = sum_{f | s}{w_f} sum_{c} {w_c g_c^2}

Here {f | s} indicate the frames in each signal block, w_f is the frame weight, w_c is the channel_weight, g_c is the channel gain, and d_{f, c} is the frame data value for frame f and channel c.

Parameters:

channel_groupChannelGroup

modeling_framesnumpy.ndarray (bool), optional

A boolean mask where True indicates that a frame is used for modeling and should be excluded from the calculations.

Returns:

gain_increment, increment_weightnumpy.ndarray, numpy.ndarray

The gain increments and associated weights both of shape (n_signal,) or (n_frames // self.resolution).

get_parms()

Return the degrees of freedom for the signal.

The degrees of freedom are given as

DOF = number_of(weights > 0) * (1 - 1/drift_n)

Returns:

float

get_robust_correlated(channel_group, modeling_frames=None, max_dependence=0.25)

Get the median derived correlated gain increment and weight.

Derived in a very similar way to the get_ml_correlated method except replacing the mean operation with a median. Please see snf.get_robust_correlated() for further details.

Parameters:

channel_groupChannelGroup

modeling_framesnumpy.ndarray (bool), optional

A boolean mask where True indicates that a frame is used for modeling and should be excluded from the calculations.

max_dependencefloat, optional

Returns:

gain_increment, increment_weightnumpy.ndarray, numpy.ndarray

The gain increments and associated weights both of shape (n_signal,) or (n_frames // self.resolution).

get_variance()

Return the signal variance.

The signal variance is returned as:

v = sum(w * x^2) / sum(w)

where x are the signal values and w are the signal weights.

Returns:

variancefloat

integrate()

Integrate the signal values and weights in-place.

Note that the spacing between sample values is assumed to be in seconds. The is_floating attribute is set to True following this operation indicating that the integrated signal has an arbitrary DC offset.

Returns:

None

level(start_frame=None, end_frame=None, robust=False)

Remove the mean value from the signal values.

The mean signal between a given start and end frame is calculated and subtracted from the signal. This value is returned to the user.

Parameters:

start_frameint, optional

The starting frame from which to level. The default is the first frame (0).

end_frameint, optional

The last frame from to level (non-inclusive). The default is the total number of frames.

robustbool, optional

If True, subtract the median value rather than the mean.

Returns:

averagefloat

The average signal value that was removed.

update(robust=False)

Update the frame data by the signal.

The gain deltas are derived and subtracted from the frame data. Dependents are updated and new source filtering are derived.

Parameters:

robustbool, optional

If True use the robust method (median) to derive means.

Returns:

None

weight_at(frame)

Return the weight at a given frame index.

Parameters:

frameint

Returns:

weightfloat