RT Journal Article
T1 PLAM: a posit logarithm-approximate multiplier
A1 Murillo Montero, Raúl
A1 Del Barrio García, Alberto Antonio
A1 Botella Juan, Guillermo
A1 Kim, Min Soo
A1 Kim, HyunJin
A1 Bagherzadeh, Nader
AB The Posit™ Number System was introduced in 2017 as a replacement for floating-point numbers. Since then, the community has explored its application in several areas, such as deep learning, and produced some unit designs which are still far from being competitive with their floating-point counterparts. This article proposes a Posit Logarithm-Approximate Multiplication (PLAM) scheme to significantly reduce the complexity of posit multipliers, one of the most power-hungry arithmetic units. The impact of this approach is evaluated in deep neural network inference, where there are no significant accuracy drops. Compared with state-of-the-art posit multipliers, experiments show that the proposed technique reduces the area, power, and delay of 32-bit hardware multipliers up to 72.86%, 81.79%, and 17.01%, respectively.
PB IEEE
SN 2168-6750
YR 2021
FD 2021-09-06
LK https://hdl.handle.net/20.500.14352/132729
UL https://hdl.handle.net/20.500.14352/132729
LA eng
NO R. Murillo, A. A. Del Barrio, G. Botella, M. S. Kim, H. Kim and N. Bagherzadeh, "PLAM: A Posit Logarithm-Approximate Multiplier," in IEEE Transactions on Emerging Topics in Computing, vol. 10, no. 4, pp. 2079-2085, 1 Oct.-Dec. 2022, doi: 10.1109/TETC.2021.3109127.
NO © 2022, IEEEPR2003_20/01
NO Fundación BBVA
NO European Commission
NO Ministerio de Ciencia e Innovación (España)
NO Agencia Estatal de Investigación (España)
NO Comunidad de Madrid
NO National Research Foundation of Korea
DS Docta Complutense
RD 10 abr 2026