Home

Omnical is a Python/C++ code set that performs redundant calibration on radio interferometer data sets. Redundant calibration is a fast and accurate algorithm that calibrate on a per frequency and per time basis, returning a set of complex gains for each antenna as well as the best fit visibilities for each unique baseline. The code's current speed is ~0.001(nAnt/32)^2 seconds/slice, on a single CPU thread. The code is applicable to any array layout with redundant baselines. While miriad is the only directly supported file format at the moment, other data formats can be easily massaged to be processed by omnical.

The code is organized and improved by Jeff Zheng (jeff_z@mit). The original contributor is the MITEoR group at MIT.