In this article, we investigate the application of blind source separation methods to extract independent components from signals recorded at the output of detectors used in γ-ray spectrometry. First, we calculate the probability density and the autocorrelation functions of each recorded signal. This allowed us to confirm that the independent component analysis algorithms, based on the computation of higher order statistics, are the best one to solve the blind source separation problem in our case of study. Among all algorithms of this approach, only four are found to be stable. The classification of these algorithms according to their performance index of separability showed that the symmetric pre-whitening algorithm is the most efficient one to achieve the separation task. Tests were performed in the presence and the absence of gamma radiation emitter.

The pulse pileup phenomenon is one of the major problems of the gamma-ray spectrometry, especially at high counting rates. This phenomenon is considered herein, in signal processing point of view, as a blind source separation problem. Indeed, we present in this paper the application of the symetric prewhitening algorithm for the separation of overlapping pulses. The obtained results show that we can achieve the separation task using only three preamplifier's recordings. In case of pulse pileup , the algorithm allows to separate the stacked pulses into two independent components. A third component is also extracted. This last was found to be strongly correlated to the detector's background noise. Otherwise, the algorithm permits to identify two independent components.

In this study, the thin independent component analysis algorithm is used to solve the blind source extraction problem in the case where the observed mixtures are defined as the HPGe preamplifier's output signals. These last correspond to the response of the detector to a combination of gamma radiation emitters having different levels of radioactivity. Indeed, on the basis of the performance index values, we conclude that this algorithm is the best blind source extraction method to analyze our data. Once the separation task is achieved, we evaluate the signal to noise ratio from individual columns of the mixing matrix. The values of this parameter permit us to detect easily the number of radionuclides used in the  Corresponding author 1158 Abdelhamid Mekaoui et al.experiment. Also, we calculate and plot the correlation functions between the signals recorded using one radioactive element and the extracted independent components. The interpretation of the gotten graphics allows us to associate each estimated independent component to the appropriate gamma radiation emitter.

Abstract. Developments in neutron detection technology during the recent past years have experienced an emphasis in their application in various fields. The performance test of one linear position sensitive 3 He detector coupled with a position decoder and multi-channel analyzer (MCA) was recorded. This system is used as the neutron powder diffractometer of CENM-Maamora. The wall effect and saturation of gas multiplication have been studied.