[Skip to Content]

Project area/S

  • Fast transients and pulsars
  • Data Intensive Astronomy

Project Details

Discovering new pulsars and exploiting them for new physics and astrophysics has a proven track record of yielding high-impact scientific dividends in the long run. Australia’s Murchison Widefield Array (MWA) provides a new parameter space in this context. One of the major goals of our Southern-sky MWA Rapid Two-metre (SMART) pulsar survey is to discover hundreds of pulsars that were missed by previous surveys. This involves processing around three Petabyes of data, and sifting through millions of candidate pulsar detections, only a small fraction of which will be genuine signals from pulsars. We are currently integrating and developing some machine-learning (ML) based software in order to efficiently identify prospective pulsar candidate signals and discriminate them from zillions of spurious ones arising from sources such as radio frequency interference.

The project involves further development and testing of the existing ML software to achieve optimum efficiency and correctness of candidate classification. Once suitable strategies are developed, the student will design and implement a user interface for running these ML tools on Garrawarla, the new supercomputer at Pawsey dedicated to processing large astronomical data sets. Finally, the whole pipeline will be tested on the candidates emerging from the SMART survey, to discover new pulsars!

Student Attributes

Academic Background

3rd or 4th year BSc (Hons.) or BEng (Hons.) student with Physics and/or astronomy track preferred, as well as those with computer science background.

Computing Skills

Python, Unix/Linux operating systems.

Training Requirement

HPC methods, parallel processing.

Project Timeline

  • Week 1 Inductions and project introduction
  • Week 2 Initial Presentation
  • Week 3 Learning to use and run the existing ML software
  • Week 4 Testing the ML software for robustness/efficiency
  • Week 5 Testing (and possibly optimising) the ML tools
  • Week 6 Identifying optimum ML strategy for the MWA searches
  • Week 7 Implementing a user interface on Garrawarla
  • Week 8 Applying pipeline to survey data, and search for new pulsars
  • Week 9 Final Presentation
  • Week 10 Final Report


Dr Ramesh Bhat

Senior Research Fellow

Read More