The High Time Resolution Universe Pulsar Survey – VI. An artificial neural network and timing of 75 pulsars

Authors

S. D. Bates, The University of ManchesterFollow
M. Bailes, Swinburne University of Technology
B. R. Barsdell, Swinburne University of TechnologyFollow
N. D. R. Bhat, Swinburne University of TechnologyFollow
M. Burgay, INAF–Osservatorio Astronomico di CagliariFollow
S. Burke-Spolaor, NASA Jet Propulsion LaboratoryFollow
D. J. Champion, MPI fuer RadioastronomieFollow
P. Coster, Swinburne University of Technology
N. D’Amico, INAF–Osservatorio Astronomico di Cagliari
A. Jameson, Swinburne University of TechnologyFollow
S. Johnston, Australia Telescope National Facility
M. J. Keith, Australia Telescope National Facility
M. Kramer, West Virginia University
L. Levin, Swinburne University of Technology
A. Lyne, West Virginia University
S. Milia, INAF–Osservatorio Astronomico di Cagliari
C. Ng, MPI fuer Radioastronomie
C. Nietner, The University of Manchester
A. Possenti, INAF–Osservatorio Astronomico di CagliariFollow
B. Stappers, The University of ManchesterFollow
D. Thornton, The University of ManchesterFollow
W. van Straten, Swinburne University of TechnologyFollow

Date of this Version

2012

Citation

Mon. Not. R. Astron. Soc. 427, 1052–1065 (2012); doi:10.1111/j.1365-2966.2012.22042.x

Abstract

We present 75 pulsars discovered in the mid-latitude portion of the High Time Resolution Universe survey, 54 of which have full timing solutions. All the pulsars have spin periods greater than 100 ms, and none of those with timing solutions is in binaries. Two display particularly interesting behaviour; PSR J1054−5944 is found to be an intermittent pulsar, and PSR J1809−0119 has glitched twice since its discovery.

In the second half of the paper we discuss the development and application of an artificial neural network in the data-processing pipeline for the survey. We discuss the tests that were used to generate scores and find that our neural network was able to reject over 99 per cent of the candidates produced in the data processing, and able to blindly detect 85 per cent of pulsars. We suggest that improvements to the accuracy should be possible if further care is taken when training an artificial neural network; for example, ensuring that a representative sample of the pulsar population is used during the training process, or the use of different artificial neural networks for the detection of different types of pulsars.