A comparison between a neural network and the likelihood method to evaluate the performance of a transition radiation detector

https://doi.org/10.1016/0010-4655(93)90139-4Get rights and content

Abstract

A classification system able to evaluate the performances of a transition radiation detector prototype for electrons/hadrons discrimination is presented. It is based both on a layered feed-forward neural network trained using back-propagation and a likehood ratio technique. The information fed into the classification system consists of the number of hits detected by each multiwires proportional chamber of the detector. The best results are obtained by the neural network approach that successfully identifies 4.0 GeV/c electrons with an hadron contamination of about 4x10-3 at 98% acceptance efficiency.

References (20)

  • B. Dolgoshein

    Nucl. Instrum. Methods A

    (1993)
  • J. Cobb

    Nucl. Instrum. Methods

    (1977)
  • R. Bellotti

    Nucl. Phys. B

    (1991)
  • E. Barbarito

    Nucl. Instrum. Methods A

    (1992)
  • C.W. Fabjan

    Nucl. Instrum. Methods

    (1981)
  • R.O. Duda et al.

    Pattern Classification and Scene Analysis

    (1973)
  • R.D. Appuhn

    Nucl. Instrum. Methods A

    (1988)
  • A. Vacchi

    Nucl. Instrum. Methods A

    (1986)
  • A. Denisov et al., preprint Fermilab-Conf-84/134-E...
  • D. Errede et al., preprint Fermilab-Conf-89/170-E...
There are more references available in the full text version of this article.

Cited by (10)

  • Sweep-Net: An Artificial Neural Network for radiation transport solves

    2021, Journal of Computational Physics
    Citation Excerpt :

    In the radiation transport community, ANNs have been seldomly used in comparison to other engineering disciplines. Prior work using Neural Networks can be found for solar irradiation spectrum reconstruction [17], to improve the radiative scalar flux approximation in space radiation sensors [18], and to reproduce one-group transport solutions in slab geometry [19] [20]. However, ANNs have not yet been applied to accelerate radiation transport problems.

  • Electron/pion identification with ALICE TRD prototypes using a neural network algorithm

    2005, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    Citation Excerpt :

    Neural networks are used for a variety of tasks in modern particle detectors [7]. A first NN analysis for electron/pion identification with a TRD [8] showed that the performance can be significantly improved. We report results for pion rejection using a NN, which increases the pion rejection factor up to about 500 for a momentum of 2 GeV/c.

View all citing articles on Scopus
View full text