Simulation study on event positioning algorithm of fast neutron imaging detector based on scintillating fiber and SiPM array

The miniaturized fast neutron imaging detector based on scintillating fiber (Sci-Fi) will adopt a SiPM array as the photoelectric conversion component. The center of gravity (CoG) method, a commonly used method for neutron event positioning, exhibits poor applicability to signals in the edge region of the array. To address this issue, this paper conducts a simulation study on the distribution characteristics of light output from the Sci-Fi and light guide using Geant4. The correlation coefficients (R2) of the photon distribution functions fitted for different regions are all above 0.99. In addition, this paper classifies the deviations generated during the application of the imaging detector into three major categories: inherent deviations, hardware deviations, and algorithmic deviations. For algorithmic deviations, scintillation photon events randomly distributed in the Sci-Fi array are simulated. The inversion method developed based on the photon distribution function adopts an iterative calculation approach to determine the event position with the best match. Compared with the CoG method, the inversion method reduces the average deviation by more than 13 %, maintains good positioning performance for events in the edge region, improves the utilization rate of the SiPM array, and is suitable for scenarios requiring higher precision and a wider field of view.