-
Notifications
You must be signed in to change notification settings - Fork 0
Home
Reports to be uploaded and/or linked to here.
Action items: Marcellin to upload code to githuib repository.
Let's adopt the following configurations:
-
JVLA 1.4 GHz (try placing the passband around the first or second null of the primary beam).
-
MeerKAT 700 MHz (passband at first null)
-
LOFAR HBA 150 MHz (passband at first null)
-
VLBI/EVN -- check with Roger Deane for a realistic scenario. Here, the primary beam will be a lot bigger than the FoV, so what we actually want to aim for is less source suppression (bigger FoV) at the same data rate.
Use NVSS or SKADS (S^3) to make a realistic sky model for the first three.
In each case the following needs to be worked out:
-
What's the critical sampling rate of each telescope (i.e. the time it takes the longest baseline to traverse 1/2 dish size in the uv-plane)
-
What is the best passband/source suppression we can achieve at that sampling rate, using e.g. 1, 10, 50 MHz frequency channels, with what WF. Produce plots of the source suppression as a function of radius, out to 20-30deg.
-
what is the compression ratio, with and without baseline-dependent averaging
-
what is the corresponding sensitivity penalty (noise ratio)
What is the fundamental relationship between degree of source suppression, passband sharpness, field of view, etc.?
Look for ways to approximate the position-variable PSF with some kind of analytic correction function. I.e. PSF(l,m;l0,m0) = PSF0(dl,dm)*F(dl,dm,l0,m0). Here, PSF0(l,m) is the nominal PSF, PSF(l,m;l0,m0) is the PSF at position l0,m0, and F is the correction function.