Advanced

Ultra low power hardware for computing squared euclidean distances

Nilsson, Peter LU and Hertz, Erik LU (2011) European Conference on Circuit Theory and Design (ECCTD 2011)
Abstract
Computing Euclidean Distances is a very important operation in digital communication, especially in the case of trellis coded modulation, where it is used numerously. This paper shows that a substantial reduction in complexity can be achieved in hardware processing elements for computing Euclidean Distances. A reduction in complexity down to 34% is shown compared to traditional designs. The paper also shows that the optimized design can be done completely ripple free, which leads to a reduction of the critical path to far more than half. The reduction in complexity leads to a reduction in power consumption. The ripple free design also leads to lower power consumption for two reasons: the rippling in itself leads to unnecessary glitches,... (More)
Computing Euclidean Distances is a very important operation in digital communication, especially in the case of trellis coded modulation, where it is used numerously. This paper shows that a substantial reduction in complexity can be achieved in hardware processing elements for computing Euclidean Distances. A reduction in complexity down to 34% is shown compared to traditional designs. The paper also shows that the optimized design can be done completely ripple free, which leads to a reduction of the critical path to far more than half. The reduction in complexity leads to a reduction in power consumption. The ripple free design also leads to lower power consumption for two reasons: the rippling in itself leads to unnecessary glitches, which costs power and the shorter critical path enables a lower supply voltage, which reduces the power consumption as well. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to conference
publication status
published
subject
conference name
European Conference on Circuit Theory and Design (ECCTD 2011)
external identifiers
  • scopus:80155192288
language
English
LU publication?
yes
id
a863a7aa-97a4-44fb-ba96-ff26232d3a24 (old id 2027683)
date added to LUP
2011-08-17 13:42:29
date last changed
2017-01-01 08:19:21
@misc{a863a7aa-97a4-44fb-ba96-ff26232d3a24,
  abstract     = {Computing Euclidean Distances is a very important operation in digital communication, especially in the case of trellis coded modulation, where it is used numerously. This paper shows that a substantial reduction in complexity can be achieved in hardware processing elements for computing Euclidean Distances. A reduction in complexity down to 34% is shown compared to traditional designs. The paper also shows that the optimized design can be done completely ripple free, which leads to a reduction of the critical path to far more than half. The reduction in complexity leads to a reduction in power consumption. The ripple free design also leads to lower power consumption for two reasons: the rippling in itself leads to unnecessary glitches, which costs power and the shorter critical path enables a lower supply voltage, which reduces the power consumption as well.},
  author       = {Nilsson, Peter and Hertz, Erik},
  language     = {eng},
  title        = {Ultra low power hardware for computing squared euclidean distances},
  year         = {2011},
}