Power-accuracy tradeoffs in human activity transition detection

Jeffrey Boyd, Hari Sundaram, Aviral Shrivastava

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Wearable, mobile computing platforms are envisioned to be used in out-patient monitoring and care. These systems continuously perform signal filtering, transformations, and classification, which are quite compute intensive, and quickly drain the system energy. The design space of these human activity sensors is large and includes the choice of sampling frequency, feature detection algorithm, length of the window of transition detection etc., and all these choices fundamentally trade-off power/performance for accuracy of detection. In this work, we explore this design space, and make several interesting conclusions that can be used as rules of thumb for quick, yet power-efficient designs of such systems. For instance, we find that the x-axis of our signal, which was oriented to be parallel to the forearm, is the most important signal to be monitored, for our set of hand activities. Our experimental results show that by carefully choosing system design parameters, there is considerable (5X) scope of improving the performance/power of the system, for minimal (5%) loss in accuracy.

Original languageEnglish (US)
Title of host publicationProceedings -Design, Automation and Test in Europe, DATE
Pages1524-1529
Number of pages6
StatePublished - 2010
EventDesign, Automation and Test in Europe Conference and Exhibition, DATE 2010 - Dresden, Germany
Duration: Mar 8 2010Mar 12 2010

Other

OtherDesign, Automation and Test in Europe Conference and Exhibition, DATE 2010
CountryGermany
CityDresden
Period3/8/103/12/10

Fingerprint

Patient monitoring
Mobile computing
Systems analysis
Sampling
Sensors

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Boyd, J., Sundaram, H., & Shrivastava, A. (2010). Power-accuracy tradeoffs in human activity transition detection. In Proceedings -Design, Automation and Test in Europe, DATE (pp. 1524-1529). [5457053]

Power-accuracy tradeoffs in human activity transition detection. / Boyd, Jeffrey; Sundaram, Hari; Shrivastava, Aviral.

Proceedings -Design, Automation and Test in Europe, DATE. 2010. p. 1524-1529 5457053.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Boyd, J, Sundaram, H & Shrivastava, A 2010, Power-accuracy tradeoffs in human activity transition detection. in Proceedings -Design, Automation and Test in Europe, DATE., 5457053, pp. 1524-1529, Design, Automation and Test in Europe Conference and Exhibition, DATE 2010, Dresden, Germany, 3/8/10.
Boyd J, Sundaram H, Shrivastava A. Power-accuracy tradeoffs in human activity transition detection. In Proceedings -Design, Automation and Test in Europe, DATE. 2010. p. 1524-1529. 5457053
Boyd, Jeffrey ; Sundaram, Hari ; Shrivastava, Aviral. / Power-accuracy tradeoffs in human activity transition detection. Proceedings -Design, Automation and Test in Europe, DATE. 2010. pp. 1524-1529
@inproceedings{fc0569c32d7b400ab77287cde43f6e48,
title = "Power-accuracy tradeoffs in human activity transition detection",
abstract = "Wearable, mobile computing platforms are envisioned to be used in out-patient monitoring and care. These systems continuously perform signal filtering, transformations, and classification, which are quite compute intensive, and quickly drain the system energy. The design space of these human activity sensors is large and includes the choice of sampling frequency, feature detection algorithm, length of the window of transition detection etc., and all these choices fundamentally trade-off power/performance for accuracy of detection. In this work, we explore this design space, and make several interesting conclusions that can be used as rules of thumb for quick, yet power-efficient designs of such systems. For instance, we find that the x-axis of our signal, which was oriented to be parallel to the forearm, is the most important signal to be monitored, for our set of hand activities. Our experimental results show that by carefully choosing system design parameters, there is considerable (5X) scope of improving the performance/power of the system, for minimal (5{\%}) loss in accuracy.",
author = "Jeffrey Boyd and Hari Sundaram and Aviral Shrivastava",
year = "2010",
language = "English (US)",
isbn = "9783981080162",
pages = "1524--1529",
booktitle = "Proceedings -Design, Automation and Test in Europe, DATE",

}

TY - GEN

T1 - Power-accuracy tradeoffs in human activity transition detection

AU - Boyd, Jeffrey

AU - Sundaram, Hari

AU - Shrivastava, Aviral

PY - 2010

Y1 - 2010

N2 - Wearable, mobile computing platforms are envisioned to be used in out-patient monitoring and care. These systems continuously perform signal filtering, transformations, and classification, which are quite compute intensive, and quickly drain the system energy. The design space of these human activity sensors is large and includes the choice of sampling frequency, feature detection algorithm, length of the window of transition detection etc., and all these choices fundamentally trade-off power/performance for accuracy of detection. In this work, we explore this design space, and make several interesting conclusions that can be used as rules of thumb for quick, yet power-efficient designs of such systems. For instance, we find that the x-axis of our signal, which was oriented to be parallel to the forearm, is the most important signal to be monitored, for our set of hand activities. Our experimental results show that by carefully choosing system design parameters, there is considerable (5X) scope of improving the performance/power of the system, for minimal (5%) loss in accuracy.

AB - Wearable, mobile computing platforms are envisioned to be used in out-patient monitoring and care. These systems continuously perform signal filtering, transformations, and classification, which are quite compute intensive, and quickly drain the system energy. The design space of these human activity sensors is large and includes the choice of sampling frequency, feature detection algorithm, length of the window of transition detection etc., and all these choices fundamentally trade-off power/performance for accuracy of detection. In this work, we explore this design space, and make several interesting conclusions that can be used as rules of thumb for quick, yet power-efficient designs of such systems. For instance, we find that the x-axis of our signal, which was oriented to be parallel to the forearm, is the most important signal to be monitored, for our set of hand activities. Our experimental results show that by carefully choosing system design parameters, there is considerable (5X) scope of improving the performance/power of the system, for minimal (5%) loss in accuracy.

UR - http://www.scopus.com/inward/record.url?scp=77953084845&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953084845&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9783981080162

SP - 1524

EP - 1529

BT - Proceedings -Design, Automation and Test in Europe, DATE

ER -