Strange things -- Open questions: Difference between revisions
Jump to navigation
Jump to search
(Created page with "'''NVis count in listobs''' * The number of visibilities corresponding to a given integration does not take into account neither the number of channels nor the number of polariz...") |
No edit summary |
| (One intermediate revision by the same user not shown) | |
(No difference)
| |
Latest revision as of 12:27, 1 September 2011
NVis count in listobs
- The number of visibilities corresponding to a given integration does not take into account neither the number of channels nor the number of polarizations. 1 spectral window and 1 baseline and 1 integration = 1 visibility
- The auto-correlations are regarded as visibilities: N antennas correspond to N*(N-1)/2 baselines and {N*(N-1)/2}+N visibilities.
Example [NGC3256]: 1 integration, 9 antennas (36 baselines), 4 spectral windows (of 128,128,1,1 channels and 2 polarizations) => NVis=180
- In the case of the WVR data (i.e. one of the spectral window), N antennas correspond to NVis=N (no correlation). And sometimes 2 separate integrations are put together in the listobs (NVis=2*N in that case). This is strange. [found in NGC3256 dataset from CSV]