TY - JOUR
T1 - Energy-dissipation limits in variance-based computing
AU - Kondapalli, Sri Harsha
AU - Zhang, Xuan
AU - Chakrabartty, Shantanu
N1 - Funding Information:
This work was supported in part by a research grant from the National Science Foundation Grant No. CNS1646380.
Publisher Copyright:
© 2018 World Scientific Publishing Company.
PY - 2018/6/1
Y1 - 2018/6/1
N2 - Variance-based logic (VBL) uses the fluctuations or the variance in the state of a particle or a physical quantity to represent different logic levels. In this paper, we show that compared to the traditional bi-stable logic representation, the variance-based representation can theoretically achieve a superior performance trade-off (in terms of energy dissipation and information capacity) when operating at fundamental limits imposed by thermal noise. We show that, in addition to the universal KT ln(1/FOR VERIFICATION) energy dissipation required for a single bit flip, a bi-stable logic device needs to dissipate at least 4.35KT/bit of energy, whereas under similar operating conditions, a VBL device reduces the additional energy dissipation requirements down to sub-KT/bit. These theoretical results are generally enough to be applicable to different instantiations and variants of VBL ranging from digital processors based on energy-scavenging or to processors based on the emerging valleytronic devices.
AB - Variance-based logic (VBL) uses the fluctuations or the variance in the state of a particle or a physical quantity to represent different logic levels. In this paper, we show that compared to the traditional bi-stable logic representation, the variance-based representation can theoretically achieve a superior performance trade-off (in terms of energy dissipation and information capacity) when operating at fundamental limits imposed by thermal noise. We show that, in addition to the universal KT ln(1/FOR VERIFICATION) energy dissipation required for a single bit flip, a bi-stable logic device needs to dissipate at least 4.35KT/bit of energy, whereas under similar operating conditions, a VBL device reduces the additional energy dissipation requirements down to sub-KT/bit. These theoretical results are generally enough to be applicable to different instantiations and variants of VBL ranging from digital processors based on energy-scavenging or to processors based on the emerging valleytronic devices.
KW - Channel capacity
KW - Dissipation limits in logic gates
KW - Energy dissipation
KW - Error probability
KW - Variance based logic
UR - http://www.scopus.com/inward/record.url?scp=85044781338&partnerID=8YFLogxK
U2 - 10.1142/S021947751850013X
DO - 10.1142/S021947751850013X
M3 - Article
AN - SCOPUS:85044781338
SN - 0219-4775
VL - 17
JO - Fluctuation and Noise Letters
JF - Fluctuation and Noise Letters
IS - 2
M1 - 1850013
ER -