A foreground model-independent Bayesian CMB temperature and polarization signal reconstruction and cosmological parameter estimation over large angular scales

Abstract

ABSTRACT Recent CMB observations have resulted in very precise observational data. A robust and reliable CMB reconstruction technique can lead to efficient estimation of the cosmological parameters. We demonstrate the performance of our methodology using simulated temperature and polarization observations using cosmic variance-limited future-generation PRISM satellite mission. We generate samples from the joint distribution by implementing the CMB inverse covariance weighted internal-linear-combination (ILC) with the Gibbs sampling technique. We use the Python Sky Model (PySM), d4f1s1 to generate the realistic foreground templates. The synchrotron emission is parametrized by a spatially varying spectral index, whereas the thermal dust emission is described as a two-component dust model. We estimate the marginalized densities of CMB signal and theoretical angular power spectrum utilizing the samples from the entire posterior distribution. The best-fitting cleaned CMB map and the corresponding angular power spectrum are consistent with the CMB realization and the sky angular power spectrum, implying an efficient foreground-minimized reconstruction. The likelihood function estimated by making use of the Blackwell–Rao estimator is used for the estimation of cosmological parameters. Our methodology can estimate the tensor-to-scalar ratio r ≥ 0.0075 for the chosen foreground models and the instrumental noise levels. Our current work demonstrates an analysis pipeline starting from the reliable estimation of CMB signal and its angular power spectrum to the case of cosmological parameter estimation using the foreground model-independent Gibbs–ILC method.