Pulse Shape Discrimination of Neutrons and Gamma Rays Using Kohonen Artificial Neural Networks

Abstract

The potential of two Kohonen artificial neural networks I ANNs) - linear vector quantisa - tion (LVQ) and the self organising map (SOM) - is explored for pulse shape discrimination (PSD), i.e. for distinguishing between neutrons (n's) and gamma rays (γ’s). The effect that la) the energy level, and lb) the relative- of the training and lest sets, have on iden- tification accuracy is also evaluated on the given PSD datasel The two Kohonen ANNs demonstrate compfcmentary discrimination ability on the training and test sets: while the LVQ is consistently mote accurate on classifying the training set. the SOM exhibits higher n/γ identification rales when classifying new paltms regardless of the proportion of training and test set patterns at the different energy levels: the average tint: for decision making equals ∼100 /e in the cax of the LVQ and ∼450 μs in the case of the SOM.

The neutron–gamma Feynman variance to mean approach: Gamma detection and total neutron–gamma detection (theory and practice)

“…Therefore, the value of the variance is deviating from the mean value of the gamma detections for both sources. It is interesting to notice that the asymptotic values of the variance to mean ratio for two 22 Na-sources are higher than for the 252 Cf source, despite the fact that in general there are more gammas emitted in a 252 Cf-source event in comparison to a 22 Na-source event. However, the strength of the 22 Na sources is at least an order of magnitude higher than the strength of 252 Cf.…”

“…the type of detectors used), since in a liquid scintillation detector each individual photon undergoes a multiple number of Compton scatterings etc., while in a theory it is assumed that the first detection removes particles from the system. One, of course, may try to assume that the ratio between the declared activity and the calculated activity should be the same for both 22 Na-sources. Such a procedure would yield the 22 Na orphan source has an activity of 18.3975 kBq.…”

“…One, of course, may try to assume that the ratio between the declared activity and the calculated activity should be the same for both 22 Na-sources. Such a procedure would yield the 22 Na orphan source has an activity of 18.3975 kBq. However, this is not a reliable approach.…”

“…The dependence of the ratio of the variance-to-mean of the number of gamma detections on the detection time for 252 Cf, 137 Cs and two22 Na sources.…”

Adaptation of RBM Learning for Intel MIC Architecture

“…Neural networks possess a natural parallelism that can be implemented on various physical architectures [3,4,5,6,7,8,37,49]. In fact, the perceptron proposed by Frank Rosenblatt [39] was implemented as a parallel electro-mechanical device.…”