Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/management/atuc/2012oct/science_meeting/docs/justin_bray_transients.pdf Дата изменения: Mon Oct 29 04:25:06 2012 Дата индексирования: Tue Feb 5 23:49:38 2013 Кодировка:

Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Transients with Parkes
What it's good at, and how to make it better

Justin Bray

29 October, 2012


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Contents

Targeted searches Blind searches
Slow transients Fast transients Really fast transients

Technical considerations


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Assumed receivers

single-pixel feed from suggestion by Dick Manchester 0.7­4.0 GHz Tsys 25 K

Phased-Array Feed (PAF) developed for ASKAP 0.7­1.8 GHz T
sys

50 K

Transients are typically less intense at higher frequencies.


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Targeted searches
Use cases: Monitoring likely/known transients Follow-up on non-radio transients Science: stellar transients (brown dwarfs, flare stars) AGN outbursts X-ray binaries intraday variability maser flares GRB follow-up Only point-source sensitivity matters: Smin = N
2kTsys 1 t Aeff obs


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Compared to other instruments
more sensitive
Arecibo

Smin (Jy) (5, 1 s)

10-3
EVLA

Parkes single-pixel
ATCA Parkes multibeam ASKAP

MeerKAT

10-2 1.0

0.5

less sensitive 0.0 sin(dec)

0.5

1.0


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Blind searches: Slow transients

Transients slower than telescope dwell time, O(5 secs) Science: blind searches for stellar transients AGN outbursts orphan GRB afterglows Figure of merit is survey speed:
2 FoV/Smin


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Compared to other instruments
fast mapping
Arecibo ALFA

Arecibo

10-3

Smin (Jy) (5, 1 s)

Parkes single-pixel
constant survey speed

EVLA MeerKAT ATCA Parkes multibeam

10-2
slow mapping

Parkes PAF

ASKAP

10-3

10-2

10-1

FoV (deg2 )

100

101

102

after Macquart et al. (2010), Cordes (2009)


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Blind searches: Fast transients
Science: RRATs radio supernovae Lorimer bursts atmospheric transients (perytons?) Cannot use standard interferometric imaging with arrays. Options: form tied-array beams sum antennas incoherently fly's-eye pointing others being explored (Bannister & Cornwell 2011) No clear figure of merit (see Macquart 2011). Search sensitivity-FoV phase space.


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Compared to other instruments
frequent weak events
Arecibo

rare weak events

10-3

Arecibo ALFA

Parkes single-pixel

Smin (Jy) (5, 1 s)

Parkes EVLA multibeam Parkes PAF ASKAP 10-2 (tied-array) (incoherent) ATCA ATCA (incoherent) (fly's eye)

10-1
frequent strong events

EVLA (fly's-eye) ASKAP (incoherent)

10-3

10-2

10-1

FoV (deg )
2

100

101

rare strong events

ASKAP (fly's eye)

102

103


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Blind searches: Really fast transients

Pulses with inherent width O(nanosecs); smaller than array sizes. Science: cosmic ray air showers lunar particle cascades nano-giant pulses Cannot use arrays together, except with tied-array beams. Signal scales with instead of (coherent across band). Figure of merit: /SEFD (and FoV)


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Compared to other instruments
frequent strong events
Arecibo

rare strong events

108

Parkes single-pixel
Arecibo ALFA

/SEFD (Hz/Jy)

Parkes PAF

107

ASKAP (tied-array)

ATCA Parkes (fly's eye) multibeam EVLA (fly's-eye)

106 105 frequent weak events 10-1 10-3 10-2
ASKAP (fly's eye)

FoV (deg )
2

100

101

102

rare weak events

103


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Triggering

real-time data

As time resolution increases, so does data rate. Cannot store all data. Need triggered detection. Need real-time processing.

stored buffer

Different from pulsar searches.


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Anticoincidence
RFI not RFI

Pol A Beam 1

Pol B Beam 1

Pol A

Pol B

Beam 2

Beam 3

Beam 4

0

1000

2000 3000 Time (ns)

0

1000

2000 3000 Time (ns)

Beam 4 0 1000

Beam 3

Beam 2

2000 3000 Time (ns)

0

1000

2000 3000 Time (ns)

One-off fast transients look much like RFI. But: they appear in multiple beams.


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Anticoincidence
105 104

After cuts:

Frames per bin

103 102 101 100 6 7

none loose extended tight expected

Peak amplitude in frame ()

8

9

10

11

12

1, 000, 000 : 0 RFI rejection is possible. But not with a single-pixel feed. Omnidirectional feed external to dish, for RFI identification?


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Conflicting technical requirements

ASKAP PAFs may be interlaced/dithered to achieve flat field. Great for imaging; not so great for transients. ASKAP PAF bands: 700­1200, 850­1440 and 1400­1800 MHz. On Parkes: separate beamformers? Makes transients difficult.


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Non-standard PAF beams

Some experiments have different pointing requirements. Lorimer bursts: reconstruction improved if beam is on target. Buffer individual PAF elements?


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Backend options

For slow transients, backend requirements are as for continuum mapping. For fast transients, options: Adaptable backend capable of implementing required processing for any conceivable experiment. No fast transients capability. Replace with the above in five years' time, when computing power is cheaper. Provide capacity for external groups to bring their own backends for particular experiments. (Where in the signal path?) Harder for a PAF than for a single-pixel feed.


Targeted searches

Slow transients

Fast transients

Really fast transients

Technical considerations

Conclusions

With new receivers, Parkes can explore new regions of transient phase space. For slow transients, arrays win. Parkes is better at fast transients. Signal-processing becomes harder as you go to shorter timescales.