A&A Supplement series, Vol. 127, March I 1998, 377-387
Received August 2, 1996; accepted July 1, 1997
M.J.D. Linden-Vørnle 
- H.U. Nørgaard-Nielsen 
Send offprint request: M.J.D. Linden-Vørnle
Niels Bohr Institute for Astronomy, Physics and Geophysics,
Astronomical Observatory, Juliane Maries Vej 30,
DK-2100 Kobenhavn O, Denmark
Danish Space Research Institute,
Juliane Maries Vej 30, DK-2100 Kobenhavn O, Denmark
The PLANCK satellite (previously known as COBRAS/SAMBA) is chosen to be the next medium sized mission in ESAs Horizon 2000 programme. It is intended to observe anisotropies in the cosmic microwave background radiation (CMBR) with a sensitivity and angular resolution which is far better than previous CMBR experiments like NASAs COBE satellite mission. In order to assess PLANCKs ability to retrieve the temperature fluctuations in the presence of contaminating foreground sources and noise, we have performed model calculations based on a single-pixel approach and a non-linear least squares spectral fitting technique.
We find that even worst-case foreground scenarios do not
prohibit a measurement accuracy of 
. This
implies that the above stated accuracy can be achieved over most of
the sky and not only in selected areas with low foreground contamination as
discussed by Brandt et al. (1994).
We have applied the model calculations to different mission configurations in order to determine the optimal combination of frequency bands. It is shown that only a large frequency coverage allows a sufficiently accurate separation of CMBR anisotropies and foreground signals. Also the Microwave Anisotropy Probe (MAP) CMBR mission, selected as one of the NASA MIDEX missions, is tested for its ability to overcome foreground contamination.
keywords: cosmic microwave background -- ISM: general -- space vehicles -- methods: data analysis