Title :
A random-search method for designing finite-wordlength recursive digital filters
Author_Institution :
University of Adelaide, Adelaide, South Australia.
fDate :
2/1/1979 12:00:00 AM
Abstract :
A new approach is described for finding the optimal finite-wordlength coefficients of recursive digital filters. A theoretical justification for the new method is presented and computational experience is reported, allowing comparison with some other methods proposed in the literature. The method appears to be suitable for designing filters of considerable complexity, using a realistic amount of computer time.
Keywords :
Approximation algorithms; Design methodology; Digital filters; Eigenvalues and eigenfunctions; Geometry; Minimax techniques; Performance evaluation; Stability criteria; Symmetric matrices; Taylor series;
Journal_Title :
Acoustics, Speech and Signal Processing, IEEE Transactions on
DOI :
10.1109/TASSP.1979.1163190
