Comparative Analysis and Optimization of Digital CR-RCn Pulse Shaping Algorithms for 3He Neutron Detectors

Abstract

Precise neutron signal discrimination and timing measurements are critical in nuclear material analysis systems utilizing Differential Die-Away (DDA) techniques, particularly under high-count-rate conditions. In this study, to ensure economic efficiency and design flexibility, a list-mode based digital data acquisition system was established by directly digitizing the preamplifier output signals of 3He proportional counters, bypassing conventional analog signal processing units such as shaping amplifiers. To mitigate the pile-up effects of preamplifier signals, CR-RCn recursive algorithms—well-suited for real-time processing—were introduced; however, systematic research on performance variations across different derivation methods and their specific suitability for neutron measurement environments remains insufficient. To address this, the algorithm proposed in this study (Case 1) and three representative existing algorithms (Case 2: Nakhostin, Case 3: Liu, and Case 4: Hong) were selected and comparatively verified through Monte Carlo simulations and neutron generator experiments. The performance was evaluated using seven metrics: universal parameters such as noise suppression, computational cost, and ballistic deficit, as well as specialized indicators essential for 3He neutron detection, including discrimination margin, full-energy peak resolution, pile-up resistance, and sampling frequency robustness. The results demonstrated that Case 2 is unsuitable for high-count-rate environments due to baseline undershoot caused by the absence of a pole-zero cancellation function. Case 4, despite its relatively superior resistance to ballistic deficit, faced limitations in real-time measurement due to excessive computation time (approximately four times longer than other algorithms) resulting from its serial processing structure. In contrast, while Case 1 and Case 3 exhibited comparable overall performance, Case 1 demonstrated overwhelming stability in sampling frequency robustness. Given that this provides a significant practical advantage by reducing the burden of recalibration caused by sampling variations during DDA system optimization, Case 1 was selected as the final algorithm for the general-purpose CPU-based DDA system in this study.