Power-aware action recognition with optimal sensor selection: an AdaBoost driven distributed template matching approach

Authors:
Pasquale Panuccio;Hassan Ghasemzadeh;Giancarlo Fortino;Roozbeh Jafari
Affiliations:
Univ. of Calabria, Rende(CS), Italy;Univ. of California, Los Angeles, CA;Univ. of Calabria, Rende(CS), Italy;Univ. of Texas at Dallas, Richardson, TX
Venue:
Proceedings of the First ACM Workshop on Mobile Systems, Applications, and Services for Healthcare
Year:
2011

Citing 6
Cited 0

Some NP-complete problems in quadratic and nonlinear programming

Mathematical Programming: Series A and B
Activity Recognition and Monitoring Using Multiple Sensors on Different Body Positions

BSN '06 Proceedings of the International Workshop on Wearable and Implantable Body Sensor Networks
Activity recognition from accelerometer data

IAAI'05 Proceedings of the 17th conference on Innovative applications of artificial intelligence - Volume 3
Energy-efficient information-driven coverage for physical movement monitoring in body sensor networks

IEEE Journal on Selected Areas in Communications - Special issue on body area networking: Technology and applications
Activity recognition from on-body sensors: accuracy-power trade-off by dynamic sensor selection

EWSN'08 Proceedings of the 5th European conference on Wireless sensor networks
A practical approach to recognizing physical activities

PERVASIVE'06 Proceedings of the 4th international conference on Pervasive Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we present a distributed action recognition framework that minimizes power consumption of the system subject to a lower bound on the classification accuracy. The system utilizes computationally simple template matching blocks that perform classifications on individual sensor nodes. A boosting approach is employed to enhance accuracy by activating only a subset of sensors optimized in terms of power consumption and can achieve a given lower bound accuracy criterion. Our experimental results on real data shows more than 85% power saving while maintaining 80% sensitivity to detected actions.