NuRadioReco.utilities.noise module
- NuRadioReco.utilities.noise.rolled_sum_roll(traces, rolling)[source]
calculates rolled sum via np.roll
- Parameters:
- traces: list
list containing 1D traces
- rolling: list
list of offsets per trace for rolling
- Returns:
- sumtr: np.array
1D array of summed traces
- NuRadioReco.utilities.noise.rolling_indices(traces, rolling)[source]
pre calculates rolling index array for rolled sum via take
- Parameters:
- traces: list
list containing (at least one) 1D trace
- rolling: list
list of offsets per trace for rolling
- NuRadioReco.utilities.noise.rolled_sum_take(traces, rolling_indices)[source]
calculates rolled sum via np.take
- Parameters:
- traces: list
list containing 1D traces
- rolling_indices: list
list of pre-calculated 1D index arrays for rolled sum
- Returns:
- sumtr: np.array
1D array of summed traces
- NuRadioReco.utilities.noise.rolled_sum_slicing(traces, rolling)[source]
calculates rolled sum via slicing
- Parameters:
- traces: list
list containing 1D traces
- rolling: list
list of offsets per trace for rolling
- Returns:
- sumtr: np.array
1D array of summed traces
- class NuRadioReco.utilities.noise.thermalNoiseGenerator(n_samples, sampling_rate, Vrms, threshold, time_coincidence, n_majority, time_coincidence_majority, n_channels, trigger_time, filt, noise_type='rayleigh', keep_full_band=False)[source]
Bases:
objectEfficient algorithms to generate thermal noise fluctuations that fulfill a high/low trigger + a majority coincidence logic (as used by ARIANNA)
- Parameters:
- n_samples: int
the number of samples of the trace
- sampling_rate: float
the sampling rate
- Vrms: float or list of floats
the RMS noise if only one value is supplied, it is assumed to be the same for all channels
- threshold: float or list of floats
the trigger threshold (assuming a symmetric high and low threshold) if only one value is supplied, it is assumed to be the same for all channels
- time_coincidence: float
the high/low coincidence time
- n_majority: int
specifies how many channels need to have a trigger
- time_coincidence_majority: float
multi channel coincidence window
- n_channels: int
number of channels to generate
- trigger_time: float
the trigger time (time when the trigger completes)
- filt: array of floats or list of array of floats
the filter that should be applied after noise generation (needs to match frequency binning) if only one array is supplied, it is assumed to be the same for all channels
- noise_type: string
the type of the noise, can be * “rayleigh” (default) * “noise”
- keep_full_band: bool
Flag to keep the full-band waveform instead of the one with the filter applied. Allows for triggering on a different (i.e. filt) band than what is returned, (default is False)
Methods
generates noise traces for all channels that will cause a high/low majority logic trigger
- class NuRadioReco.utilities.noise.thermalNoiseGeneratorPhasedArray(detector_filename, station_id, triggered_channels, Vrms, threshold, ref_index, noise_type='rayleigh', log_level=0, pre_trigger_time=100, trace_length=512, filt=None, upsampling=2, window_length=16, step_size=8, main_low_angle=np.float64(-1.039338088856046), main_high_angle=np.float64(1.039338088856046), n_beams=11, quantize=True)[source]
Bases:
objectEfficient algorithms to generate thermal noise fluctuations that fulfill a phased array trigger
- Parameters:
- detector_filename: string
the filename to the detector description
- station_id: int
the station id of the station from the detector file
- triggered_channels: array of ints
list of channel ids that are part of the trigger
- Vrms: float
the RMS noise
- threshold: float
the trigger threshold (assuming a symmetric high and low threshold)
- ref_index: float
reference refractive index for calculating time delays of the beams
Methods
generate_noise([phasing_mode, trigger_mode, ...])generates noise traces for all channels that will cause a high/low majority logic trigger
generate_noise2([debug])generates noise traces for all channels that will cause a high/low majority logic trigger This implementation is slow due to nested python loops each rolling the trace over and over again
- Other Parameters:
- noise_type: string
the type of the noise, can be * “rayleigh” (default) * “noise”
- log_level: logging enum, default warn
the print level for this module
- pre_trigger_time: float, default 100 ns
the time in the trace before the trigger happens
- trace_length: float, default 512 ns
the total trace length
- filt: array of complex values, default None
the filter that should be applied after noise generation (needs to match frequency binning in upsampled domain) if None, a default filter is calculated from 96 to 220 MHz
- upsampling: int, default 2
factor by which the waveforms will be upsampled before calculating time delays and beamforming
- window_length: float, default 16 ns
time interval of the integration window
- step_size: float, default 8 ns
duration of a stride between window calcuations
- main_low_angle: float, default -59.5 deg
angle (radians) of the lowest beam
- main_high_angle: float 59.5 deg
angle (radians) of the highest beam
- n_beams: int, default 11
number of beams to calculate
- quantize: bool, default True
If set to true, the conversion to and from ADC will be performed to mimic digitizations
- generate_noise(phasing_mode='slice', trigger_mode='binned_sum', debug=False)[source]
generates noise traces for all channels that will cause a high/low majority logic trigger
- Parameters:
- phasing_mode: string (default: “slice”)
“slice” or “roll”, two implementations for phasing by either slicing the array or using np.roll. The default shows better performance, but “roll” is kept as an alternative
- trigger_mode: string (default: “binned_sum”)
“binned_sum” or “stride”, two implementations for triggering The default shows better performance, but “stride” is kept as an alternative
- debug:
generate debug plot
- Returns
- ——-
- np.array of shape (n_channels, n_samples), index of triggered bin, index of triggered beam