pycrtools.tasks.crosscorrelateantennas

CrossCorrelateAntennas([ws, parfile])	Usage:
FindPulseDelay([ws, parfile])	Calculate pulse delay by finding maxima in a trace.
Task([ws, parfile])	Base class from which all tasks should be derived.

Module documentation

Module author: Heino Falcke <h.falcke@astro.ru.nl>

class pycrtools.tasks.crosscorrelateantennas.CrossCorrelateAntennas(ws=None, parfile=None, **kwargs)

Usage:

Task = trun("CrossCorrelateAntennas", timeseries_data[n_dataset, blocksize], refant=0) -> Task.crosscorr_data (i.e., strength of cross-correlation as a function of time)

To provide only fft data use:

Task = trun("CrossCorrelateAntennas", fft_data[n_dataset, blocksize / 2 + 1], refant=0) -> Task.crosscorr_data (i.e., strength of cross-correlation as a function of time)

Description:

Calculates the cross-correlation of a number of time series data sets with respective to a reference data set.

The data set is a matrix of N equal-length timeseries. If not reference data set is provided, a reference antenna from the input time series matrix is specified (by default = 0, i.e. all data is correlated with the first data set, then the first cross correlation output will actually be an autocorrelation).

Make sure N is even. This is not tested for uneven N.

NB: This has also not yet been tested for NyquistZone 2!

If timeseries_data == None, then one needs to provide fft_data instead and the fft will not be (re-)calculated. Note that the function uses fftw convention!

Input parameters

blocksize

Length of the data for each antenna

crosscorr_data_orig

Scratch cross correlation vector in original size to hold intermediate results

fft_data

FFT of the input timeseries data

fft_reference_data

FFT of the reference data, dimensions [N data sets, (data length)/2+1]. If no extra reference antenna array is provided, cross correlate with reference antenna in data set

fftwplan

Memory and plan for FFT

ifftwplan

Memory and plan for inverse FFT

oversamplefactor [default value: 1 ]

Oversample the final cross-correlation by so many sub-steps. If equal to one then no oversampling.

refant [default value: 0 ]

Which data set in data to use as reference data set if no reference_data provided.

reference_data [default value: None ]

Reference data set to cross-correlate data with. If None then use data from reference antenna refant.

timeseries_data [default value: None ]

Timeseries data, if None fft_data must be given.

Output parameters

crosscorr_data

Output array of dimensions [N data sets, data length * oversamplefactor] containing the cross correlation.

class pycrtools.tasks.crosscorrelateantennas.FindPulseDelay(ws=None, parfile=None, **kwargs)

Calculate pulse delay by finding maxima in a trace.

This may require some additional manipulation to the data to give a meaningfull result, for instance the trace may be the cross correlation signal.

Input parameters

plot_prefix [default value: “” ]

Prefix for plots

plot_type [default value: “png” ]

Plot type (e.g. png, jpeg, pdf)

plotlist [default value: [] ]

List of plots

refant [default value: None ]

reference antenna

sampling_frequency [default value: 200000000.0 ]

sampling frequency in Hz.

save_plots [default value: False ]

Store plots

trace [default value: None ]

array containing a trace as a function of time for each antenna

Output parameters

delays

pulse delays in seconds.

maxpos

position of pulse maximum.