|
|||
NRAO Home > CASA > CASA Toolkit Reference Manual |
|
calibrater.setsolvegainspline - Function
2.1.1 Specialization of setsolve for cubic spline G (time-dependent gain)
solving
Description
This function is a specialization of the setsolve method which should be used
when cubic spline G solutions are desired, e.g., when SNR on calibrators is
very low. Currently, this solving mode treats dual polarization data on a
per-polarization basis. The option to obtain a joint solution (a la ’T’) will be
provided in the future.
The visibility data are averaged in frequency (for multi-channel data) prior to the solution.
This method uses many of the basic parameters as the generic setsolve. Parameters unique to the spline solver are:
- mode
- For phase solutions only, use mode=’PHAS’. For amplitude solutions only, use mode=’AMP’. If both are desired, use mode=’PHASAMP’, and both will be solved for using the same spline timescale (this mode also assumes that all calibrators have the correct relative flux densities). If solving for phase and amplitude separately (usually in this order), it is usually desirable to apply the first one when solving for the second one. Spline solution so obtained will be stored in separate calibration tables. In the near future, the mode parameter will be consolidated with the generic apmode parameter.
- splinetime
- The spline timescale (time between knots) is specified here. The default is 10800 seconds (3 hours). In future this parameter will be consolidated with the generic t parameter. The preavg parameter should be set to a value at least 4X shorter than the spline time (an error will occur if there is insufficient sampling within the splinetime timescale), and consistent with the expected coherence. Consistent with these constraints, use the largest possible value for preavg to optimize the SNR of the pre-solve phase-tracking algorithm.
- npointaver and phasewrap
- These parameters tune the phase-unwrapping algorithm when mode = ’PHAS’. Cycle slips are detected (and removed before the spline solve) when the median phase a sequence of length npointaver (in integrations) differs by more than phasewrap degrees from the previous sequence.
Pending improvements:
- Consolidate more parameters with the generic setsolve
- Introduce the generic combine options
- Improve phase-tracking algorithm
Arguments
Inputs |
| ||
table |
| Output calibration table name
| |
| allowed: | string |
|
| Default: |
| |
append |
| Append to existing table?
| |
| allowed: | bool | |
| Default: | false |
|
mode |
| Phase or Amplitude mode?
| |
| allowed: | string |
|
| Default: | AMP PHASAMP PHAS |
|
splinetime |
| Spline timescale (sec)
| |
| allowed: | double |
|
| Default: | 10800 |
|
preavg |
| Pre-averaging interval (in sec)
| |
| allowed: | double |
|
| Default: | 0.0 |
|
npointaver |
| ||
| allowed: | int |
|
| Default: | 10 |
|
phasewrap |
| ||
| allowed: | double |
|
| Default: | 250 |
|
refant |
| Reference antenna. Default is none.
| |
| allowed: | any |
|
| Default: | variant
|
|
bool
Example
cb.open(’ngc5921.ms’)
cb.selectvis(field=’1445*’)
cb.setsolvegainspline (table=’gcalph’,mode=’PHAS’,splinetime=3600.0,preavg=60.0)cb.solve()
cb.setsolvegainspline (table=’gcalamp’,mode=’AMP’,splinetime=10800.0);
cb.solve();
cb.close();
In this example, a spline solution is first found for phase on a hourly timescale, then for amplitude on a three-hour timescale.
__________________________________________________________________
More information about CASA may be found at the
CASA web page
Copyright © 2016 Associated Universities Inc., Washington, D.C.
This code is available under the terms of the GNU General Public Lincense
Home |
Contact Us |
Directories |
Site Map |
Help |
Privacy Policy |
Search