Optimum synthesis of FIR filters with arbitrary time and frequency constraints for energy and time estimations in case of pulse-correlated noise

摘要

We present two noticeable extensions to the applications of the DPLMS (Digital Penalized Least Mean Squares) method for optimum energy filter synthesis that we introduced a few years ago. By using the DPLMS method, high performance digital filters with time constraints, e.g. flat-top, finite duration and also frequency constraint can be derived from easily acquired experimental signals. Now, we want to show i) how the DPLMS method can be used to synthesize also optimum digital filters for time estimations of signals and ii) how the DPLMS method can be directly used also in presence of time variant, pulse correlated noise, e.g. in case of scintillation detectors coupled to photomultiplier tubes or avalanche photodiodes, for both energy and time estimations. Experimental validation of the DPLMS method for both energy and time estimations of signals coming from LaBr scintillator detectors is provided.