Source separation in astrophysical maps using independent factor analysis
Neural Networks - 2003 Special issue: Neural network analysis of complex scientific data: Astronomy and geosciences
| Hi-index | 0.00 |
The source separation problem applied to the electromagnetic maps from the PLANCK satellite measurements is attacked here by means of an Independent Component Analysis (ICA) approach. This kind of techniques are totally blind with respect to the system to be identified, and only assume statistical independence of the signals to be separated. The computational scheme of our ICA implementation is described and compared with other existing separation strategies, and some results from simulated experiments are shown. TEL:: +39 050 3153137 EMAIL:: salerno@iei.pi.cnr.it