Simulation

This module generates synthetic MeerKAT-like time-ordered data for testing and validation. It replicates the pipeline from Zhang et al. (2026, §4): azimuth raster scan → equatorial coordinate conversion → Gaussian beam projection → flicker noise generation.

The primary class is TODSimulation, which runs the full simulation on construction and stores all components as attributes.

Key attributes of TODSimulation

Attribute	Description
TOD_setting	Observed time-ordered data (gain × Tsys × (1 + noise))
sky_params	True sky pixel temperatures (HEALPix)
gains_setting	True gain time series
noise_setting	True \(1/f\) noise realisation
time_list	Observation times (seconds from scan start)
gain_proj	Legendre projection matrix for the gain model
Tsky_operator	Beam projection matrix mapping sky pixels → TOD samples
Tloc_operator	Local-temperature projection matrix (noise diode + receiver)
logfc, logf0, alpha	True noise parameters
wnoise_var	White-noise variance \(\sigma_w^2\)

MultiTODSimulation combines a TODSimulation in setting mode and one in rising mode, providing the two overlapping scans used in the 2×TOD analysis of Zhang et al. (2026).