Calibrating photon-counting detectors to high accuracy: background and deadtime issues

Abstract

Abstract When photon-counting detectors are calibrated in the presence of a background signal, deadtime effects can be significant and must be carefully accounted for to achieve high accuracy. We present a method for separating the correlated signal from the background signal that appropriately handles deadtime effects. This method includes consideration of pulse timing and afterpulsing issues that arise in typical avalanche photodiode (APD) detectors. We illustrate how these effects should be accounted for in the calibration process. We also discuss detector timing issues that should be considered in detector calibration. Acknowledgement This work was supported in part by DTO, ARO and DARPA/QUIST. Notes †The specification that all correlated signal occurs after d bins is, of course, artificial and introduced only for notational convenience. In practice, it is usually easier just to record the correlated signal and then pad the leading portion of the histogram with some extra background rather than using extra long DUT delays. †To introduce a simple correction for afterpulses due to uncorrelated events, one could assume that the number of afterpulses in the histogram bin i is proportional to the number of events in the bin (i − d) with a constant probability of an afterpulse β i.e. by adding βH(τ i − d ) to the right-hand sides of equations (Equation6) and (Equation7); β can be determined as a ratio of peaks D and A.