Development of a procedure and design of an instrument to correct for pulse pile-up distortion

Several areas of science and medicine involve the collection of physical measurements in pulse form where the information is carried by the pulse height. If the pulse rate is high enough, then two or more randomly-spaced pulses may occur closely spaced. If this spacing is sufficiently close, the skirts of the pulse waveshapes will overlap, producing false pulse heights. Current pulse height instruments handle this problem by rejecting the closely-spaced pulses and then increasing experiment time to compensate. It would be desirable, instead, to be able to instrumentally analyze the pulses and correct for the error caused by pile-up. In this thesis, a mathematical development of the nature of pile-up distortion is performed which shows that the pile-up effects for a set of N pulses can be described with a set of N simultaneous linear equations. The parameters and coefficients of the equation set are generated from the false pulse heights and the pulse spacing. The algorithm was tested in a programmable calculator, and then an instrument was designed that could perform on-line pulse pile-up correction.