Upper limits on by direct experimental measurements on ground, in the range, have been performed in the past. The first limit by ground-based gravitational wave detectors was , set by a coincidence observation between bar detectors in Glasgow (Hough et al. 1975). More recent estimates from the background noise of cryogenic resonant bar detectors of the Rome group (Astone et al. 1996) give . The value obtained with the ALTAIR detector is . With only one detector it is possible to give an upper limit only, and in order to improve the sensitivity we need at least two detectors. The problem of crosscorrelating gravitational wave detectors to search for a stochastic background has been discussed by a number of authors. We recall (Vitale et al. 1997) that the background appears as a noise in each detector, but if the detectors are in the same location and orientation then the effect of the background will be perfectly correlated between them. Detectors that are separated by more than, roughly, a wave-length will be only partly correlated, since waves from different directions excite them with different time delays. The analysis of the data of the 100-hour experiment, obtained with the Glasgow and Garching interferometers, for a preliminary 6 minutes of data (Compton et al. 1996) gives a value for . Obviously all these values of do not set interesting limits on the stochastic background, but they give us the opportunity to develop adeguate tecniques of data analysis. In fact these upper limits have to be compared with the nucleosynthesis limit (Brustein et al. 1995) of at all scales of frequency.
© European Southern Observatory (ESO) 1999
Online publication: March 1, 1999