New classes of astronomical objects are often discovered serendipitously. The enormous data volumes produced by recent high-time resolution, radio-telescope surveys imply that efficient algorithms are required for a discovery. Such algorithms are usually tuned to detect specific, known sources. Existing data sets therefore likely contain unknown astronomical sources, which will remain undetected unless algorithms are developed that can detect a more diverse range of signals. We present the Single-dish PARKES data challenge for finding the uneXpected (SPARKESX), a compilation of real and simulated high-time resolution observations. SPARKESX comprises three mock surveys from the Parkes"Murriyang"radio telescope. A broad selection of simulated and injected expected signals (such as pulsars, fast radio bursts), poorly characterised signals (plausible flare star signatures) and unknown unknowns are generated for each survey. The goal of this challenge is to aid in the development of new algorithms that can detect a wide-range of source types. We show how successful a typical pipeline based on the standard pulsar search software, PRESTO, is at finding the injected signals. The dataset is publicly available at https://doi.org/10.25919/fd4f-0g20.
[1]
Yunqian Ma,et al.
Imbalanced Learning: Foundations, Algorithms, and Applications
,
2013
.
[2]
D. R. Lorimer,et al.
SIGPROC: Pulsar Signal Processing Programs
,
2011
.
[3]
William D. Pence,et al.
CFITSIO: A FITS File Subroutine Library
,
2010
.
[4]
R. Lathe.
Phd by thesis
,
1988,
Nature.
[5]
Ewen Barr,et al.
Peasoup: C++/CUDA GPU pulsar searching library
,
2020
.
[6]
N. RayP.RugglesNorrisCliveL..
Accelerating the Rate of Astronomical Discovery
,
2010
.
[7]
Ronald D. Ekers,et al.
SETI 2020 : a roadmap for the search for extraterrestrial intelligence / produced for the SETI Institute by the SETI Science & Technology Working Group
,
2002
.