Enhancing the trapezoid filter into a Sliding Least Squares fitter for the Nab DAQ

The trapezoid filter is ubiquitous in nuclear spectroscopy due to its ability to handle ballistic deficit and exponential background, using a recursive implementation. However, offline pulse processing often involves least squares fitting to template waveforms for better energy resolution and a Chi squared goodness-of-fit. We have developed a sliding least squares fitter (SLS) as an enhancement to the trapezoid filter, which is able to fit to an exponential tail pulse and provide Chi squared in real time. Implementation of the SLS will enable more efficient noise and signal differentiation, which will reduce data storage requirements.