Effects of high count rate and gain shift on isotope-identification algorithms

Radio-Isotope Identification Algorithms for NaI γ Spectra

“…For example [3] and [16] showed that a typical peak centroid can shift by tens of keV in either direction. If in a given spectum, all peak centroids shift in the same direction, then corrections such as those in [16] can be effective. Nevertheless, such adjustments cannot be expected to be perfect.…”

“…If instead two repeated measurements of the same source were made using the same detector and model, we expect peak centroid locations shifts to be in the same direction. One option to align such shifted spectra requires making multiple measurements of the source, each from a different detector location to map the effect of high count rate (for small source-detector distances) on calibration-shiftrelated peak centroid shifting [16].…”

“…The published literature on RIID algorithms using NaI spectra, much of which is referenced here, tends to isolate one or a few noise sources and evaluate mitigation strategies, such as the peak alignment strategy described in [16]. Therefore, in conjunction with the Bayesian variable selection approach, a combination of real and simulated spectra should be used in extensive testing to assess the individual impact of all noise sources and potential noise mitigation strategies as measurement protocols are varied.…”