Statistics – Computation
Scientific paper
Dec 1996
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=1996aas...18910701b&link_type=abstract
American Astronomical Society, 189th AAS Meeting, #107.01; Bulletin of the American Astronomical Society, Vol. 28, p.1417
Statistics
Computation
Scientific paper
Recent exciting advances in wide-field imaging of low-frequency radio interferometry data have had significant scientific impact on a number of important astrophysical problems. Thermal-noise or confusion-limited images from 330 MHz or 74 MHz data taken with the Very Large Array (VLA) often resemble optical photographs in appearance. Fields as large as 8 degrees on a side may be obtained at a single pointing. Routine use of these data has been limited by the computational burden of properly imaging the fields at full resolution. The low-frequency radio sky is filled with sources and one must image the whole primary beam or more to remove the effect of confusing sources. Also, the usual two-dimensional approximation used in Fourier-synthesis imaging breaks down on these wide fields. The program dragon exists for the solution of the wide-field problem but is very slow. In the common ``contiguous faceting'' mode of operation, it completely tessellates the curved sky with small tangent planes. We have ported dragon to the parallel architecture SGI Power Challenge Array and can now produce VLA B-configuration images in a few hours. For high-resolution A-configuration images, still extremely computationally challenging, we have implemented a ``targeted faceting'' approach to sidelobe removal. A low-resolution image is constructed and used to automatically tabulate all sources in the primary beam. A large but tractable region of the sky is tessellated with normal contiguous facets, and the remaining sources are addressed with small targeted facets centered on each region of emission. Treatment of outlying sources is similar to the NRAO program MX, but a complete deconvolution to thermal noise may require 200 to 800 targeted facets instead of a few. The resulting high-resolution image may be a degree or more in size, while still reducing the computational load by an order of magnitude. Full-resolution images of the supernova remnant/HII complex W49 will be shown to illustrate the power of this method.
Briggs D. S.
Foster Roger S.
Kassim Namir E.
No associations
LandOfFree
Recent Advances in Low-Radio-Frequency Wide-Field Imaging does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Recent Advances in Low-Radio-Frequency Wide-Field Imaging, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Recent Advances in Low-Radio-Frequency Wide-Field Imaging will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-818253