Component separation for Cosmic Microwave Background data: a blind approach based on spectral diversity

We present a blind multi-detector multi-component spectral matching method for all sky observations of the cosmic microwave background, working on the spherical harmonics basis. The method allows to estimate on a set of observation maps the power spectra of various components present in CMB data, their contribution levels in each detector and the noise levels. The method accounts for the instrumental effect of beam convolution. We have implemented the method on all sky Planck simulations containing five components and white noise, including beam smoothing effects.