TY - GEN
T1 - Energy efficient activity recognition based on low resolution accelerometer in smart phones
AU - Liang, Yunji
AU - Zhou, Xingshe
AU - Yu, Zhiwen
AU - Guo, Bin
AU - Yang, Yue
PY - 2012
Y1 - 2012
N2 - Smart phone is becoming an ideal platform for continuous and transparent sensing with lots of built-in sensors. Activity recognition on smart phones is still a challenge due to the constraints of resources, such as battery lifetime, computational workload. Keeping in view the demand of low energy activity recognition for mobile devices, we propose an energy-efficient method to recognize user activities based on a single low resolution tri-axial accelerometer in smart phones. This paper presents a hierarchical recognition scheme with variable step size, which reduces the cost of time consuming frequency domain features for low energy consumption and adjusts the size of sliding window to improve the recognition accuracy. Experimental results demonstrate the effectiveness of the proposed algorithm with more than 85% recognition accuracy for 11 activities and 3.2 hours extended battery life for mobile phones.
AB - Smart phone is becoming an ideal platform for continuous and transparent sensing with lots of built-in sensors. Activity recognition on smart phones is still a challenge due to the constraints of resources, such as battery lifetime, computational workload. Keeping in view the demand of low energy activity recognition for mobile devices, we propose an energy-efficient method to recognize user activities based on a single low resolution tri-axial accelerometer in smart phones. This paper presents a hierarchical recognition scheme with variable step size, which reduces the cost of time consuming frequency domain features for low energy consumption and adjusts the size of sliding window to improve the recognition accuracy. Experimental results demonstrate the effectiveness of the proposed algorithm with more than 85% recognition accuracy for 11 activities and 3.2 hours extended battery life for mobile phones.
KW - activity recognition
KW - energy efficient
KW - hierarchical recognition
KW - low resolution
KW - tri-axial accelerometer
UR - http://www.scopus.com/inward/record.url?scp=84861129034&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-30767-6_11
DO - 10.1007/978-3-642-30767-6_11
M3 - 会议稿件
AN - SCOPUS:84861129034
SN - 9783642307669
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 122
EP - 136
BT - Advances in Grid and Pervasive Computing - 7th International Conference, GPC 2012, Proceedings
T2 - 7th International Conference on Advances in Grid and Pervasive Computing, GPC 2012
Y2 - 11 May 2012 through 13 May 2012
ER -
