Transient Radio Sky

Theme description

We live in an ocean of electromagnetic waves of both manmade and natural origin. Electromagnetic waves of natural origin carry important information about both the terrestrial and the cosmological environments. These waves have been detected and studied at all ranges of the electromagnetic spectrum, from radio up to Υ-rays. Most these signals have either continuous, repetitive, or of predictable time-domain behaviour. The techniques developed to detect such waves typically deliberately avoid unpredictable transient signals. This means that natural phenomena that emit short or irregular bursts of electromagnetic radiation, such as solar flares or cosmic ray induced air showers, remain little studied and poorly understood. Such events are considered as an important source of new science.

 Exploring the transient radio sky is one of the major science drivers for several upcoming and planned radio telescopes, including Square Kilometer Array (SKA). The detection of transient signals, particularly very short durations, can however be considerably harmed by the prevalent man made radio frequency interferences (RFI), unless multiple detectors are located on the remote sites on which RFIs do not correlate to each other.

New digital radio technologies, combined with new high-speed networks and high performance computing have created a unique opportunity to develop instruments to explore the phenomena of fast transients.

To this end, we have been developing a Transient Radio Events Sensor Network for deployment in New Zealand since 2006. This sensor network will allow us to register and statistically study transient radio events in the 15-20 MHz frequency band where some of the event may have sufficient brightness to be detected with a relatively basic antenna.

The most essential part of this network is the Digital Receiver Sensor (DRS) which has been developed at the Centre for Reconfigurable Systems of AUT. The DRS is based around a Field Programmable Gate Array (FPGA) which has been configured as a highly specialized microcomputer for sampling and remote logging data. DRS is equipped with a high accuracy GPS that allows synchronization of all devices in the network with 25ns accuracy.

Each DRS produces up to 10MB of raw data each second. Initially data is buffered in the DRS’s memory and is subsequently uploaded onto a remote server via KAREN network. A network of 10 devices, in their maximum capability, will generate more than 8.6TB of data daily. These data is to be transported in near real-time, managed, stored and processed.

Initial 5 DRSs devices will be deployed across New Zealand by the end of 2010. The data will be continuously streamed to 8TB high speed storage allocated on NZ DataGrid, processed daily on the HPC resources participating in NZ BestGrid, first of all allocated 128 CPUs on BlueFern of UoC as well as AUT’s and UoO’s research HPC clusters. Resource, operational and middleware support of the KAREN, BestGrid and BlueFern will be essential for success of the project.

The data will be available via KAREN to all partners and the wider research community for analysis.

Theme Members

• Dr Slava Kitaev, Centre for Reconfigurable Systems, AUT
• Dr Tim Molteno, Electronics Research, Department of Physics, University of Otago
• Dr Melanie Johnston-Hollitt, Victoria University of Wellington
• Mr Achala Perera, School of Engineering, AUT

Graduate Students

·         Ian Scott (University of Otago)

There are currently 3 scholarships available for one year research based Master degrees at AUT, UoO and VUW. Enquiries are via email slava.kitaev@aut.ac.nz.

The initial stage of the project has been supported by the KAREN CBF grant for 2009-2010.

International Engagement

Part of this work has been recorded officially as part of the PrepSKA workpackage. It is the only SKARD activity so far to be officially included.