The Infrared Background due to Unresolved Sources
A peculiarity of galaxy spectra in the mm and sub-mm region is the very steep rise to shorter wavelengths following the strong spectral dependence of dust emissivity. This is the only spectral region where such a steep spectral shape is observed. The effect of the redshift on the location of the peak implies that it is possible to select high-redshift galaxies at a given flux level.
Figure 1.19: Extragalactic background in the FIR/submm/mm region. The predictions of two evolutionary schemes (PLE, pure luminosity evolution and DLE, density and luminosity evolution) fitting the IRAS data are respectively given by the solid line and the dashes. The predictions for a model of evolving radiosources (RG) is drawn with a dotted line. The Cosmic Microwave Background is also shown (CBR). Various observational upper limits are shown, as well as the predocted extragalactic background in the NIR (solid line) and the predicted Galactic starlight (dots). The solid squares show the isotropic component measured by Puget et al. (1996) from FIRAS observations.
Figure 1.19 compares the diffuse background for the 2
extreme models of figure 1.18, together with other
observational information about the FIR/sub-mm/mm backgrounds. We see
that the isotropic sub-mm component found by Puget et al. (1996) in the COBE/FIRAS data
(Hauser et al. 1998,
Fixsen et al. 1998)
is fitted by a PLE model, where the bulk of star-formation activity is confined to z 
3. In contrast, galaxy formation is assumed to be a continuous process in the DLE model, with giant galaxies forming at z~ 2. As a result of the decrease in the number density of galaxies with increasing redshift, and of the absence of cold dust in the ISM, the predicted sub-mm background is much smaller.
Recently, the FIR background detected by COBE FIRAS and DIRBE instruments has been confirmed by the ISOPHOT instrument aboard the Infrared Space Observatory (ISO). The so-called FIRBACK deep survey carried out by ISOPHOT found a source density for objects with a flux above 200 mJy that exceeds the counts expected for sources detected in the IRAS deep surveys with a similar flux by about an order of magnitude (see Puget et al. 1999).