TOAST is a software framework for simulating and processing timestream data collected by telescopes. Telescopes which collect data as timestreams rather than images give us a unique set of analysis challenges. Detector data usually contains noise which is correlated in time as well as sources of correlated signal from the instrument and the environment. Large pieces of data must often be analyzed simultaneously to extract an estimate of the sky signal.

See the full documentation here:

https://toast-cmb.readthedocs.io/en/latest/