A User-Oriented Approach to Designing FIR Deconvolution Filters

摘要

Based on learning in input-output signal domain and controlling noise amplification by varying sampling rate, a pragmatic approach is developed for designing FIR deconvolution filters with the desired noise gains producing accurate as possible output time-domain waveforms for filter support intervals defined by a user. As an example, FIR linear phase digital differentiator is designed by the developed methodology for so-called logarithmic derivative technique and its performance is compared with that of the appropriate equiripple minimax and maximally linear differentiators.