Removing Periodic Transients

Some data have periodic transients (instrument relevels, disk drive writing…) that can interfere with signal processing. The PeriodicTransient class lets you calculate and remove these transients, if they are truly periodic and always have the same shape. This is a labor-intensive, manual module, so if you use it on a dataset, please share your results! Properties and methods are:

Constructor

PeriodicTransient

Properties

• Input by the constructor

  • name (str): name of this periodic transient (e.g., ‘hourly’)

  • period (float): seconds between each transient

  • dp (float): how many seconds to change the period by when testing for better values

  • clips (tuple): clip values outside of this range (low, high). Should include the max range of the transient

  • transient_starttime (UTCDateTime`): onset time of earliest transient.

• Calculated by PeriodicTransient.calc_transient()

  • transient_model (): Model of the periodic transient.

  • dirac_comb (): offsets of transients from the first one, in seconds

  • n_transients_used (): number of transients used to make the model

  • tm (): transient starting 1 sample earlier

  • ``tp``(): transient starting 1 sample later

Methods

• calc_timing: Interactively calculate transient parameters

• calc_transient: Calculate the shape of the transient

• remove_transient: Remove the transient from a data trace

Example

tiskitpy.PeriodicTransient_example