timeDownsample

class timeDownsample : public kotekan::Stage

Average a set number of frames on a single-frequency stream to effectively reduce the cadence of the acquisition.

This stage accumulates and averages a specified number of incoming frames on a single-frequency stream to reduce the cadence of the acquisition. Visibilities, eigenvectors, eigenvalues, eigen-rms are averaged. Inverse weights are averaged and divided by number of combined frames to track reduction in variance. Metadata from the first frame is passed on and that of the others discarded. Will throw an exception if more than one frequency is found in the stream.

  • kotekan_timedownsample_skipped_frame_total The number of frames skipped entirely because they were too old.

Buffers

  • in_buf The kotekan buffer of the incoming single-frequency stream.

    • Format: VisBuffer structured

    • Metadata: VisMetadata

  • out_buf The kotekan buffer into which low cadence stream is fed.

    • Format: VisBuffer structured

    • Metadata: VisMetadata

Author

Tristan Pinsonneault-Marotte

Param num_samples:

Int. The number of time frames to average.

Param max_age:

Float. How old can a frame be in seconds before it is dropped. Default is 120 (i.e. two minutes).

Public Functions

timeDownsample(kotekan::Config &config, const std::string &unique_name, kotekan::bufferContainer &buffer_container)

Default constructor, loads config params.

virtual void main_thread() override

Main loop for the stage.