Energy-Efficient Integration of Continuous Context Sensing and Prediction into Smartwatches

被引:41
作者
Rawassizadeh, Reza [1 ]
Tomitsch, Martin [2 ]
Nourizadeh, Manouchehr [3 ]
Momeni, Elaheh [4 ]
Peery, Aaron [1 ]
Ulanova, Liudmila [1 ]
Pazzani, Michael [1 ]
机构
[1] Univ Calif Riverside, Dept Comp Sci & Engn, Riverside, CA 92521 USA
[2] Univ Sydney, Design Lab, Sydney, NSW 2006, Australia
[3] Vienna Univ Technol, A-1040 Vienna, Austria
[4] Univ Vienna, Multimedia Informat Syst Grp, A-1090 Vienna, Austria
关键词
wearable; smartwatch; mobile sensing; prediction; energy efficiency; lifelogging; quantified self; MOBILE; RECOGNITION;
D O I
10.3390/s150922616
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
As the availability and use of wearables increases, they are becoming a promising platform for context sensing and context analysis. Smartwatches are a particularly interesting platform for this purpose, as they offer salient advantages, such as their proximity to the human body. However, they also have limitations associated with their small form factor, such as processing power and battery life, which makes it difficult to simply transfer smartphone-based context sensing and prediction models to smartwatches. In this paper, we introduce an energy-efficient, generic, integrated framework for continuous context sensing and prediction on smartwatches. Our work extends previous approaches for context sensing and prediction on wrist-mounted wearables that perform predictive analytics outside the device. We offer a generic sensing module and a novel energy-efficient, on-device prediction module that is based on a semantic abstraction approach to convert sensor data into meaningful information objects, similar to human perception of a behavior. Through six evaluations, we analyze the energy efficiency of our framework modules, identify the optimal file structure for data access and demonstrate an increase in accuracy of prediction through our semantic abstraction method. The proposed framework is hardware independent and can serve as a reference model for implementing context sensing and prediction on small wearable devices beyond smartwatches, such as body-mounted cameras.
引用
收藏
页码:22616 / 22645
页数:30
相关论文
共 56 条
[1]  
[Anonymous], P SEMANTICS VIENN AU
[2]  
[Anonymous], 2012, P 21 INT C WORLD WID, DOI [10.1145/2187836.2187868, DOI 10.1145/2187836.2187868]
[3]  
[Anonymous], 1999, P SIGCHI C HUM FACT, DOI DOI 10.1145/302979.303126
[4]  
Balan RajeshKrishna., 2014, 2014 Sixth International Conference on Communication Systems and Networks (COMSNETS), P1
[5]   Data Mining for Wearable Sensors in Health Monitoring Systems: A Review of Recent Trends and Challenges [J].
Banaee, Hadi ;
Ahmed, Mobyen Uddin ;
Loutfi, Amy .
SENSORS, 2013, 13 (12) :17472-17500
[6]   Accelerating Dynamic Time Warping Clustering with a Novel Admissible Pruning Strategy [J].
Begum, Nurjahan ;
Ulanova, Liudmila ;
Wang, Jun ;
Keogh, Eamonn .
KDD'15: PROCEEDINGS OF THE 21ST ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2015, :49-58
[7]  
Berlin E., 2015, Sensors Applications Symposium (SAS), 2015 IEEE, P1
[8]  
Berlin E, 2012, UBICOMP'12: PROCEEDINGS OF THE 2012 ACM INTERNATIONAL CONFERENCE ON UBIQUITOUS COMPUTING, P250
[9]   A survey of context modelling and reasoning techniques [J].
Bettini, Claudio ;
Brdiczka, Oliver ;
Henricksen, Karen ;
Indulska, Jadwiga ;
Nicklas, Daniela ;
Ranganathan, Anand ;
Riboni, Daniele .
PERVASIVE AND MOBILE COMPUTING, 2010, 6 (02) :161-180
[10]  
Bhargava P., 2014, P 11 INT C MOB UB SY, P40, DOI DOI 10.4108/ICSTM0BIQUIT0US2014257654