Use of multiple number representation in automatic arithmetic data-path design - Archive ouverte HAL
Communication Dans Un Congrès Année : 2003

Use of multiple number representation in automatic arithmetic data-path design

Résumé

We present in this paper a study of a framework for a FIR filter with fixed coefficients targeting CMOS 0.35´ for binary standard and redundant representations. Booth recoded and Lefevre constant multipliers are used. The internal data coding being both borrow-save and carry-save, the framework manages the generation of adapted adders.
Fichier non déposé

Dates et versions

hal-01529999 , version 1 (31-05-2017)

Identifiants

Citer

Raphaël Daouphars, Laurent-Stéphane Didier. Use of multiple number representation in automatic arithmetic data-path design. Advanced Signal Processing Algorithms, Architectures, and Implementations XIII, 2003, San Diego, California, United States. pp.482-489, ⟨10.1117/12.505595⟩. ⟨hal-01529999⟩
45 Consultations
0 Téléchargements

Altmetric

Partager

More