FPGA implementation of MLP, 1D-CNN and TTTratio algorithms for neutron/gamma-ray discrimination using plastic scintillator

Ali Hachem (1) , Yoann Moline (2) , Gwenolé Corre (2) , Frédérick Carrel (2) , Imane Belachheb (2)

1 LCIM (CEA, LIST) - Laboratoire Capteurs et Instrumentation pour la Mesure (CEA, LIST)
2 LEMA (CEA, LIST) - Laboratoire architecture Electronique, Modélisation et Analyse de données (CEA, LIST)

Pulse shape discrimination algorithms, such as Tail-to-Total integral ratio (TTTratio) have been commonly integrated on edge devices for online neutron/gamma discrimination using organic scintillators. These algorithms have a number of limitations, especially with plastic scintillators which have low intrinsic discriminating ability. Machine learning (ML) models have recently been explored as a way to improve discriminating performance. Most of these methods are proposed for liquid and stilbene scintillators and do not address the embedded implementation. The purpose of this study is to implement on FPGA TTTratio, Multi Layer Perceptron Neural Network (MLP) and 1D Convolution Neural Network (1D-CNN) models trained for neutron/gamma-ray discrimination using EJ276 plastic scintillator, with a latency less than the signal duration (500 ns) using minimal resources. Therefore, the comparison between the different methods can be done according to the discrimination performance, latency and resource consumption.

Mots clés: signal processing; Neutron-gamma discrimination; ionizing radiation; radioactivity; instrumentation; nuclear instrumentation; spectrum analysis; organic scintillator; machine learning; neural network; plastic scintillator; artificial intelligence

Domaines: Electronique; Instrumentations et Détecteurs [physics.ins-det]; Machine Learning [stat.ML]

Ali Hachem, Yoann Moline, Gwenolé Corre, Frédérick Carrel, Imane Belachheb. FPGA implementation of MLP, 1D-CNN and TTTratio algorithms for neutron/gamma-ray discrimination using plastic scintillator. NorCAS 2023 - 2023 IEEE Nordic Circuits and Systems Conference, Oct 2023, Aalborg, Denmark. ⟨10.1109/NorCAS58970.2023.10305446⟩. ⟨cea-04408832⟩