Cosmic microwave background power spectrum estimation with the destriping technique

Extraction of the cosmic microwave background (CMB) angular power spectrum is a challenging task for current and future CMB experiments due to the large data sets involved. Here we describe an implementation of Monte Carlo apodized spherical transform estimator (MASTER) described in Hivon et al., which exploits the destriping technique as a map-making method. In this method a noise estimate based on destriped noise-only Monte Carlo (MC) simulations is subtracted from the pseudo-angular power spectrum. As a working case we use realistic simulations of the Planck low-frequency instrument (LFI). We found that the effect of destriping on a pure sky signal is minimal and requires no correction. Instead we found an effect related to the distribution of detector pointings, which affects the high-part of the power spectrum. We correct for this by subtracting a 'signal bias' estimated by MC simulations. We also give analytical estimates for this signal bias. Our method is fast and accurate enough (the estimator is unbiased and errors are close to theoretical expectations for maximal accuracy) to estimate the CMB angular power spectra for current and future CMB space missions. This study is related to Planck LFI activities.