|
| This Article | ||
| ||
| Share | ||
| Bibliographic References | ||
| Add to: | ||
| | ||
| Search | ||
| ||
15th IEEE Symposium on Computer Arithmetic (ARITH-15 '01)
Unrestricted Faithful Rounding is Good Enough for Some LNS Applications
Vail, Colorado
June 11-June 13
ISBN: 0-7695-1150-3
| ASCII Text | x | ||
| Mark G. Arnold, Colin Walter, "Unrestricted Faithful Rounding is Good Enough for Some LNS Applications," Computer Arithmetic, IEEE Symposium on, pp. 0237, 15th IEEE Symposium on Computer Arithmetic (ARITH-15 '01), 2001. | |||
| BibTex | x | ||
| @article{ 10.1109/ARITH.2001.930125, author = {Mark G. Arnold and Colin Walter}, title = {Unrestricted Faithful Rounding is Good Enough for Some LNS Applications}, journal ={Computer Arithmetic, IEEE Symposium on}, volume = {0}, year = {2001}, isbn = {0-7695-1150-3}, pages = {0237}, doi = {http://doi.ieeecomputersociety.org/10.1109/ARITH.2001.930125}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, } | |||
| RefWorks Procite/RefMan/Endnote | x | ||
| TY - CONF JO - Computer Arithmetic, IEEE Symposium on TI - Unrestricted Faithful Rounding is Good Enough for Some LNS Applications SN - 0-7695-1150-3 SP EP A1 - Mark G. Arnold, A1 - Colin Walter, PY - 2001 VL - 0 JA - Computer Arithmetic, IEEE Symposium on ER - | |||
Abstract: We propose relaxing the restricted form of faithful rounding used in prior 32-bit Logarithmic Number System (LNS) implementations. Unrestricted faithful rounding yields a three- to six-fold savings in VLSI ROM size (or four- to six- fold savings in F GPA table size) with only modest increase in error. This can be acceptable for the DSP and multimedia applications in which the non-standard LNS is a candidate for adoption. Such applications are cost sensitive, and the tremendous cost reduction of the LNS model proposed here should argue very favourably for its adoption.
Citation:
Mark G. Arnold, Colin Walter, "Unrestricted Faithful Rounding is Good Enough for Some LNS Applications," arith, pp.0237, 15th IEEE Symposium on Computer Arithmetic (ARITH-15 '01), 2001
Usage of this product signifies your acceptance of the Terms of Use.
