An isotope classifier of cosmic-ray events collected by space detectors has been implemented using a multi-layer perceptron neural architecture. In order to handle a great number of different isotopes a modular architecture of the “mixture of experts” type is proposed. The performance of this classifier has been tested on simulated data and has been compared with a “classical” classifying procedure. The quantitative comparison with traditional techniques shows that the neural approach has classification performances comparable – within – with that of the classical one, with efficiency of the order of . A possible hardware implementation of such a kind of neural architecture in future space missions is considered.
Cosmic Ray discrimination capabilities of ΔE-E silicon nuclear telescopes using neural network
Castellano, M.;
2000-01-01
Abstract
An isotope classifier of cosmic-ray events collected by space detectors has been implemented using a multi-layer perceptron neural architecture. In order to handle a great number of different isotopes a modular architecture of the “mixture of experts” type is proposed. The performance of this classifier has been tested on simulated data and has been compared with a “classical” classifying procedure. The quantitative comparison with traditional techniques shows that the neural approach has classification performances comparable – within – with that of the classical one, with efficiency of the order of . A possible hardware implementation of such a kind of neural architecture in future space missions is considered.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.