ANALYTICAL CORRECTIONS FOR DEAD TIME EFFECTS IN THE MEASUREMENT OF TIME-INTERVAL DISTRIBUTIONS

Corrections for the effects of instrumental dead time in the measurement of time-interval distributions are presented for the idealized cases of paralyzable and nonparalyzable detection systems. It is assumed that signal pulses occurring within the measurement cycle following each initial stimulus applied to the system under study, referred to as the source, are random and uncorrelated. Usually the total number n of pulses from the source in each measurement cycle is also random and is given by a Poisson distribution. However, correction equations are also developed for those cases where the source distribution S(n) may be represented by an arbitrary but known probability distribution. An example of their application in the field of time-of-flight mass spectrometry, where the distribution consists of peaks corresponding to different masses, is given. The limitation of the method arising from the increase in statistical noise at high count rates is analyzed.