We cross-correlate WMAP and ROSAT diffuse X-ray background maps and look for common features in both data sets. We use the power spectrum of the product maps and the cross-power spectrum to highlight a possible correlation. The power spectrum of the product maps does not detect any correlation and the cross-power spectrum does not show any significant deviation from 0. We explore different explanations for this lack of correlation. A universe with a low value of Sigma_8 could naturally explain the lack of correlation. We also discuss the systematic effects which can affect this result, in particular the subtraction of some cluster signal from the ROSAT diffuse maps which could significantly suppress the correlation signal. These systematic effects reduce considerably the significance of our constraints on the cosmological model. When we include the systematic effects we find a weaker constrain on Sigma_8 allowing models with values as large as Sigma_8=1 (for Omega_m = 0.3) to be consistent with the lack of correlation. To illustrate the capabilities of the method with future high-quality data, we show how from the correlation signal it should be possible to predict the level of contamination of the SZ effect on the power spectrum of the CMB. Within the systematic errors we find evidence that this contribution is negligible for WMAP and is expected to be small in experiments like ACBAR or CBI, but can be important for future high resolution experiments.

We cross-correlate the recent CMB WMAP 1 year data with the diffuse soft X-ray background map of ROSAT. We look for common signatures due to galaxy clusters (SZ effect in CMB, bremsstrahlung in X-rays) by cross-correlating the two maps in real and in Fourier space. We do not find any significant correlation and we explore the different reasons for this lack of correlation. The most likely candidates are the possibility that we live in a low $\sigma _8$ universe ($\sigma_8 < 0.9$) and/or systematic effects in the data especially in the diffuse X-ray maps which may suffer from significant cluster signal subtraction during the point source removal process.

We use the X-ray power spectrum of the ROSAT all-sky survey in the R6 band (approximately 0.9-1.3 keV) to set an upper limit on the galaxy cluster power spectrum. The cluster power spectrum is modelled with a minimum number of robust assumptions regarding the structure of the clusters. The power spectrum of ROSAT sets an upper limit on the Omega_m-sigma_8 plane which excludes all the models with sigma_8 above sigma_8 = 0.5/(Omega_m^0.38) in a flat LCDM universe. We discuss the possible sources of systematic errors in our conclusions, mainly dominated by the assumed L_x-T relation. Alternatively, this relation could be constrained by using the X-ray power spectrum, if the cosmological model is known. Our conclusions suggest that only models with a low value of sigma_8 (sigma_8 < 0.8 for Omega_m = 0.3) may be compatible with our upper limit. We also find that models predicting lower luminosities in galaxy clusters are favoured. Reconciling our cosmological constraints with these arising by other methods might require either a high entropy floor or wide-spread presence of cooling flows in the low-redshift clusters.

Fluctuations of the soft X-ray background are investigated using harmonic analysis. A section of the ROSAT All-Sky Survey around the north galactic pole is used. The flux distribution is expanded into a set of harmonic functions and the power spectrum is determined. Several subsamples of the RASS have been used and the spectra for different regions and energies are presented. The effects of the data binning in pixels are assessed and taken into account. The spectra of the analyzed samples reflect both small scale effects generated by strong discrete sources and the large scale gradients of the XRB distribution. Our results show that the power spectrum technique can be effectively used to investigate anisotropy of the XRB at various scales. This statistics will become a useful tool in the investigation of various XRB components.