Bochum 1998 – scientific programme
Parts | Days | Selection | Search | Downloads | Help
HK: Hadronen und Kerne
HK 13: Instrumentation II, Data Acquisition Systems
HK 13.3: Talk
Monday, March 16, 1998, 17:30–17:45, A
Classification of particles measured with a leadglass detector using a neural network — •S. Joerrens1, C. Barlag1, S. Bathe1, C. Blume1, Z. B"or"ocz1, D. Bucher1, H. B"usching1, R. Glasow1, S. Gorus1, H. K"ohncke1, T. Kr"umpel1, T. Peitzmann1, K. Reygers1, R. Santo1, H. Schlagheck1, and WA982 — 1Institut f"ur Kernphysik, Universit"at M"unster, D-48149 M"unster — 2
The LEDA (LEadglass Detector Array) from the WA98 experiment at CERN is used for the detection of photons. The main purpose of the detector is the search for direct photons from a QGP (Quark Gluon Plasma) by measuring inclusive photons and reconstructing neutral pions. For this task a good discrimination between photons and charged hadrons is mandatory. The specific response of the detector to different particles can be used for this discrimination.
In the WA98 experiment the dispersion cut, which uses the lateral
shower spread, has been established as standard method.
Studies of a neural network (NN) in comparison to the dispersion cut
have been performed with simulated GEANT data and real WA98
data. The main idea for the development of the NN is the search for
an alternvative analysis algorithm. These two analysis methods should
improve the understanding of systematic errors.
* Partly supported by the German BMBF and the GSI.