Presampling Filtering

Abstract

Sampled data systems often employ lumped-parameter lowpass filters both prior to and following the sampling operation. The purpose of these filters is to reduce the error between the input and output data waveforms. The present paper discusses the effect of presampling filters on the rms interpolation error for two types of sampled data systems and gives some thumb rules for choosing such filters. The two types of sampled data systems considered are: (1) one employing only zero-order hold interpolation, and (2) one employing zero-order hold followed by the best lowpass lumped-parameter interpolation filter. The resulting expressions for rms interpolation error for sampled data systems employing lumped-parameter filters from a detailed time domain analysis are given.