Extracting cosmic microwave background polarisation from satellite astrophysical maps

We present the application of the Fast Independent Component Analysis (FastICA) technique for blind component separation to polarised astrophysical emission. We study how the Cosmic Microwave Background (CMB) polarised signal, consisting of E and B modes, can be extracted from maps affected by substantial contamination from diffuse Galactic foregrounds and instrumental noise. We perform the analysis of all sky maps simulated accordingly to the nominal performances of the Low Frequency Instrument (LFI) aboard the Planck satellite; the sky signal is modeled as a superposition of CMB, generated by a Gaussian, nearly scale invariant cosmological perturbation spectrum, and the existing simulated polarisation templates of Galactic synchrotron. Our results indicate that the angular power spectrum of CMB E modes can be recovered on all scales up to l ≃ 1000, corresponding to the fourth acoustic oscillation, while B modes can be detected, up to their turnover at l ≃ 100 if cosmological tensor amplitude exceeds a 30% fraction of scalar perturbations. Moreover, we assess the instrumental capability to recover the cross correlation between total intensity and polarisation, TE, recovering the spectrum up to l ≃ 1200, corresponding to the seventh TE acoustic oscillation. For comparison, we study the performance of the Microwave Anisotropy Probe (MAP) satellite for polarised radiation and in presence of foregrounds; we find CMB polarisation detection in the interval 100 ≤ l ≤ 500 for TE, corresponding to the first three acoustic oscillations, and a marginal detection of the peak of the second E mode acoustic oscillation, in the interval 300 ≤ l ≤ 400.