Development of pulse shape algorithms for discrimination between gamma and X-ray radiation pulses

Area of Science:

• Nuclear instrumentation
• Signal processing
• Spectroscopy

Background:

• Gamma spectrometers can be influenced by X-ray pulses, necessitating pulse shape discrimination (PSD).
• Accurate differentiation between gamma and X-ray pulses is crucial for reliable spectrometer applications.

Purpose of the Study:

• To propose and evaluate algorithms for eliminating X-ray pulse influence in gamma spectrometers.
• To compare the performance of various PSD algorithms based on accuracy, speed, and environmental robustness.

Main Methods:

• Algorithms evaluated include curve fitting (CF), artificial neural network (ANN), system identification, peak shape, amplitude search with curve fitting, and pulse tracking.
• Data acquired using NaI(Tl) and Si(Li) detectors for gamma and X-ray pulses, respectively.
• Performance metrics included residuals, fitting error, and processing time.

Main Results:

• The peak shape algorithm demonstrated superior computational speed.
• The ANN algorithm, utilizing Bispectrum features, achieved 100% accuracy in noisy environments.
• The system identification algorithm yielded zero fitting error in clean environments.

Conclusions:

• Developed algorithms effectively differentiate gamma and X-ray pulses for improved spectrometer design.
• Algorithm selection depends on application-specific requirements for speed, accuracy, and environmental conditions.
• These PSD techniques enhance spectrometer applicability across diverse operational settings.