sxyu/watplot

Interactive waterfall plots for Breakthrough Listen data

watplot: Interactive Waterfall Plots

© Alex Yu 2019

This is a work in progress.

Overview

A interactive waterfall plot visualizer for Breakthrough Listen data, supporting dragging, zooming, and adjusting colormaps on-the-fly for small to moderately large (1-2GB) files. Currently, .fil (sigproc filterbank) files and .h5/.hdf5 (HDF5) files are supported.

Build

Dependencies:

• OpenCV 3.3+
  • sudo apt-get install libopencv-dev on Ubuntu
• HDF5 from https://www.hdfgroup.org/downloads/
  • sudo apt-get install libhdf5-dev on Ubuntu
• Eigen 3 (included in repo)

Build: Use cmake on all platforms: mkdir build && cd build && cmake ..

Then use make -j4 to build. On Windows, you may double-click the sln file generated to open Visual Studio to build. You might also need to specify the installation directory for OpenCV, by passing -DOpenCV_dir='path' to cmake.

Usage

On the Breakthrough Listen computing cluster: watplot file [f_start[%] f_stop[%] [t_start[%] t_end[%]]]

Where

• file may be a Filterbank (.fil) or HDF5 (.h5) data file in the Breakthrough Listen format
• f_start, f_end specify the frequency range, default everything
• t_start, t_end specify the time range, default everything
• Append % after any frequency or time value to use a percent of the data instead of specifying the explicit values.
  • For example, watplot file.fil 0% 50% loads the lower half of the frequencies

GUI Controls:

• Left click and drag mouse OR use WASD to pan
• To zoom, use:
  • The scroll wheel OR
  • -,= keys OR
  • Control + Shift + Left click and drag vertically
• Control + Left click and drag to zoom asymmetrically
  • Drag horizontally for frequency, vertically for time
• Press 0 to reset zoom and position
• Right click and drag mouse to adjust colormap (horizontal: shift, vertical: scale)
• Press C, Shift + C to cycle through colormaps (15 total)
• Press L to toggle log scale
• Press Shift + S to save plot to ./waterfall-NUM.png
• Press Q or ESC to exit

Sample data (Voyager): https://storage.googleapis.com/gbt_fil/voyager_f1032192_t300_v2.fil

Try zooming into Voyager's transmitting frequency, 8420.15 MHz, and you should see a powerful signal.

License

Apache 2.0.