A new twelve-transistor approximate 4:2 compressor in CNTFET technology

Power consumption is a serious concern in the field of digital design. Reducing power supply voltage, power gating, transistor downscaling, voltage over scaling, applying modern technology and approximate computing are some candidate means in reducing power consumption. Among these candidates, appro...

Full description

Saved in:
Bibliographic Details
Published in:International journal of electronics 2019-05, Vol.106 (5), p.691-706
Main Authors: Shirinabadi Farahani, Samira, Reshadinezhad, Mohammad Reza
Format: Article
Language:English
Subjects:
CMOS
Compressors
Computation
Electric potential
Moore's law
Multimedia
Multipliers
Power consumption
Power efficiency
Power supplies
Transistors
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Power consumption is a serious concern in the field of digital design. Reducing power supply voltage, power gating, transistor downscaling, voltage over scaling, applying modern technology and approximate computing are some candidate means in reducing power consumption. Among these candidates, approximate computing can generate a trade-off between accuracy and power-delay-area efficiency in error resilient applications. According to Moore's law together with CMOS problems in nanoscale regime, modern technologies emerge to solve these problems. Among these recent technologies, CNTFET technology is considered as promising. As multiplication is frequently applied in multimedia processing, implementing efficient multipliers constitute critical. Compressors are fundamental elements in reduction tree multipliers and improve their efficiency, thus an improvement in multipliers' performance. A new 12-transistor approximate 4:2 compressor is proposed here. This new appropriate compressor, in terms of area, power consumption, accuracy and reliability design, is more efficient than its existing counterparts.
ISSN:0020-7217
1362-3060
DOI:10.1080/00207217.2018.1545930