A&A Supplement series, Vol. 129, May I 1998, 647-655
Received June 5; accepted October 20, 1997
F. Bocchino - M. Barbera - S. Sciortino
Send offprint request: F. Bocchino
Istituto ed Osservatorio Astronomico "G.S. Vaiana",
Piazza del Parlamento 1, I-90134 Palermo, Italy
e-mail: bocchino@oapa.astropa.unipa.it,
barbera@oapa.astropa.unipa.it,
sciorti@oapa.astropa.unipa.it
We have developed a model-independent time screening optimization
algorithm to cope with significant contamination spikes in the
ROSAT PSPC/HRI bacground light-curves. The rejection criteria are
based on the maximization of faint sources signal-to-noise ratio. The
algorithm tuning parameters have been optimized through performing a
wide set of runs on both simulated and real data. We have verified that
the application of our selection criteria to the case of long exposure
PSPC observations yields an increase of the number of faint sources
(SNR 
) of up to 100% with a rejection of up to the 8% of the
exposure time. At the same time, we obtain an average signal-to-noise
ratio gain of 3% for the sources detected in both the unscreened and
screened images. The algorithm, even though currently optimized for
the instrumentation on board ROSAT, can be easily generalized for
data analysis of other imaging X-ray detectors.
keywords: methods: data analysis -- techniques: miscellaneous