schemes.Chen et al.[10]analyze the power consumption simple strategy can be designed to avoid the possible delay of AMOLED displays in multimedia applications.Camera for changing modes recoding incurs high power cost.LiKamWa R et al.[11]do re- search on the image sensor and reveal two energy-proportional ACKNOWLEDGMENT mechanisms which are supported by current image sensor but unused in mobile system.It indeed saves energy.But it only This work is supported in part by National Natural Science focuses on the energy consumption of the moment of shooting Foundation of China under Grant Nos.61472185,61373129. while overlooking the consumption of preparation.Han et al. 61321491,91218302;Key Project of Jiangsu Research Pro- [12]study the energy cost made by human-screen interaction gram under Grant No.BE2013116;EU FP7 IRSES Mobile- such as scrolling on screen.They propose a scrolling-speed- Cloud Project under Grant No.612212.This work is par- adaptive frame rate controlling system to save energy.Dietrich tially supported by Collaborative Innovation Center of Novel et al.[3]detect the game's current state and lower the proces- Software Technology and Industrialization.Lei Xie is the sor's voltage and frequency whenever possible to save energy. corresponding author. B.Activity Sensing REFERENCES With the development of phone's built-in sensors,various [1]"KS Mobile,"http://www.cmcm.com/zh-cn/. approaches of activity recognition have been done.They can [21 "Monsoon PowerMonitor,"https://www.msoon.com/LabEquipment/ PowerMonitor/. be classified into single-sensor and multi-sensors sensing. 31 B.Dietrich and S.Chakraborty,"Power management using game state Single sensor is used in the following work.Built-in detection on android smartphones,"in Proc.of ACM MobiSys,2013. microphone is used to detect the events that are closely related [4] X.Fan.W.-D.Weber,and L.A.Barroso,"Power provisioning for a to sleep quality,including body movement,couch and snore warehouse-sized computer,"in Proc.of ACM SIGARCH,2007. [13].Using built-in accelerometers,user's daily activities such [ F.Bellosa,A.Weissel,M.Waitz,and S.Kellner,"Event-driven energy as walking,jogging,upstairs,downstairs,sitting and standing accounting for dynamic thermal management,"in Proc.of COLP,2003. are recognized in [14].With the labeled accelerometer data. [6 D.Rajan,R.Zuck,and C.Poellabauer,"Workload-aware dual-speed dynamic voltage scaling."in Proc.of IEEE RTCSA,2006. they apply four machine learning algorithms and make some 7] N.Balasubramanian,A.Balasubramanian,and A.Venkataramani analysis.Lee et al.[15]use accelerometers with hierarchical "Energy consumption in mobile phones:a measurement study and hidden markov models to distinguish the daily actions. implications for network applications."in Proc.of ACM SIGCOMM. 2009. Multi-sensors are used in the following work.Shahzad et [8]M.Dong and L.Zhong."Self-constructive high-rate system energy al.[16]propose a gesture based user authentication scheme modeling for battery-powered mobile systems,"in Proc.of ACM for the secure unlocking of touch screen devices.It makes MobiSys,2011. use of the coordinates of each touch point on the screen, [9]F.Xu,Y.Liu,Q.Li,and Y.Zhang."V-edge:Fast self-constructive accelerometer values and time stamps.Chen et al.[17]take power modeling of smartphones based on battery voltage dynamics."in advantage of features as light,phone situation,stationary and Proc.of NSDI,2013. silence to monitor user's sleep.They need to use several [1O]X.Chen.Y.Chen,Z.Ma.and F.C.Fernandes."How is energy different sensors to obtain all the phone's information.Driving consumed in smartphone display applications?in Proc.of ACM HotMobile.2013. style,which is concerned with man's life,is recognized by [11]R.LiKamWa,B.Priyantha,M.Philipose,L.Zhong.and P.Bahl, using the gyroscope,accelerometer,magnetometer,GPS and "Energy characterization and optimization of image sensing toward video[18].Bo et al.[19]propose a framework to verify whether continuous mobile vision,"in Proc.of ACM MobiSys,2013. the current user is the legitimate owner of the smartphone [12 H.Han,J.Yu,H.Zhu,Y.Chen,J.Yang,G.Xue,Y.Zhu,and M.Li,"E based on the behavioral biometrics,including touch behaviors 3:energy-efficient engine for frame rate adaptation on smartphones,"in and walking pattens.These features are extracted from smart- Proc.of ACM Sensys.2013. phone's built-in accelerometer and gyroscope. [13]T.Hao.G.Xing,and G.Zhou,"isleep:unobtrusive sleep quality monitoring using smartphones,"in Proc.of ACM Sensys.2013. [14]J.R.Kwapisz,G.M.Weiss,and S.A.Moore,"Activity recognition VI.CONCLUSION AND FUTURE WORK using cell phone accelerometers."S/GKDD.vol.12,no.2.pp.74-82. 2011. In this paper,we propose a context aware energy-saving scheme for smart camera phone based on activity sensing. [15]Y.-S.Lee and S.-B.Cho,"Activity recognition using hierarchical hidden We take advantage of the features of activities and maintain markov models on a smartphone with 3d accelerometer,"in Proc.of Springer HAIS,2011. an activity state machine to do the recognition.Then energy [16]M.Shahzad,A.X.Liu,and A.Samuel,"Secure unlocking of mobile saving scheme is applied based on the result of recognition. touch screen devices by simple gestures:You can see it but you can Our solution can perceive the user's activities with an average not do it,"in Proc.of ACM MobiCom,2013. accuracy of 95.5%and reduce the overall energy consumption [17]Z.Chen,M.Lin,F.Chen,N.D.Lane,G.Cardone,R.Wang.T.Li, by 46.5%for smart camera phones. Y.Chen,T.Choudhury.and A.T.Campbell,"Unobtrusive sleep monitoring using smartphones,"in Proc.of IEEE PervasiveHealth Following the current research,considering the difference 2013 between users,there are three possible directions for future [18]D.A.Johnson and M.M.Trivedi,"Driving style recognition using a work.First,more data of the process can be collected with smartphone as a sensor platform,"in Proc.of IEEE ITSC,2011. our work to improve the design and implementation.Second. [19]C.Bo,L.Zhang.X.-Y.Li,Q.Huang.and Y.Wang,"Silentsense:silent a self-constructive user preference learning can be designed to user identification via touch and movement behavioral biometrics,"in Proc.of ACM MobiCom,2013 automatically extract the user perference of software settings. Third,to the phone whose configuration is too low,more 12schemes. Chen et al. [10] analyze the power consumption of AMOLED displays in multimedia applications. Camera recoding incurs high power cost. LiKamWa R et al. [11] do re￾search on the image sensor and reveal two energy-proportional mechanisms which are supported by current image sensor but unused in mobile system. It indeed saves energy. But it only focuses on the energy consumption of the moment of shooting, while overlooking the consumption of preparation. Han et al. [12] study the energy cost made by human-screen interaction such as scrolling on screen. They propose a scrolling-speed￾adaptive frame rate controlling system to save energy. Dietrich et al. [3] detect the game’s current state and lower the proces￾sor’s voltage and frequency whenever possible to save energy. B. Activity Sensing With the development of phone’s built-in sensors, various approaches of activity recognition have been done. They can be classified into single-sensor and multi-sensors sensing. Single sensor is used in the following work. Built-in microphone is used to detect the events that are closely related to sleep quality, including body movement, couch and snore [13]. Using built-in accelerometers, user’s daily activities such as walking, jogging, upstairs, downstairs, sitting and standing are recognized in [14]. With the labeled accelerometer data, they apply four machine learning algorithms and make some analysis. Lee et al. [15] use accelerometers with hierarchical hidden markov models to distinguish the daily actions. Multi-sensors are used in the following work. Shahzad et al. [16] propose a gesture based user authentication scheme for the secure unlocking of touch screen devices. It makes use of the coordinates of each touch point on the screen, accelerometer values and time stamps. Chen et al. [17] take advantage of features as light, phone situation, stationary and silence to monitor user’s sleep. They need to use several different sensors to obtain all the phone’s information. Driving style, which is concerned with man’s life, is recognized by using the gyroscope, accelerometer, magnetometer, GPS and video[18]. Bo et al. [19] propose a framework to verify whether the current user is the legitimate owner of the smartphone based on the behavioral biometrics, including touch behaviors and walking pattens. These features are extracted from smart￾phone’s built-in accelerometer and gyroscope. VI. CONCLUSION AND FUTURE WORK In this paper, we propose a context aware energy-saving scheme for smart camera phone based on activity sensing. We take advantage of the features of activities and maintain an activity state machine to do the recognition. Then energy saving scheme is applied based on the result of recognition. Our solution can perceive the user’s activities with an average accuracy of 95.5% and reduce the overall energy consumption by 46.5% for smart camera phones. Following the current research, considering the difference between users, there are three possible directions for future work. First, more data of the process can be collected with our work to improve the design and implementation. Second, a self-constructive user preference learning can be designed to automatically extract the user perference of software settings. Third,to the phone whose configuration is too low, more simple strategy can be designed to avoid the possible delay for changing modes. ACKNOWLEDGMENT This work is supported in part by National Natural Science Foundation of China under Grant Nos. 61472185,61373129, 61321491, 91218302; Key Project of Jiangsu Research Pro￾gram under Grant No. BE2013116; EU FP7 IRSES Mobile￾Cloud Project under Grant No. 612212. This work is par￾tially supported by Collaborative Innovation Center of Novel Software Technology and Industrialization. Lei Xie is the corresponding author. REFERENCES [1] “KS Mobile,” http://www.cmcm.com/zh-cn/. [2] “Monsoon PowerMonitor,” https://www.msoon.com/LabEquipment/ PowerMonitor/. [3] B. Dietrich and S. Chakraborty, “Power management using game state detection on android smartphones,” in Proc. of ACM MobiSys, 2013. [4] X. Fan, W.-D. Weber, and L. A. Barroso, “Power provisioning for a warehouse-sized computer,” in Proc. of ACM SIGARCH, 2007. [5] F. Bellosa, A. Weissel, M. Waitz, and S. Kellner, “Event-driven energy accounting for dynamic thermal management,” in Proc. of COLP, 2003. [6] D. Rajan, R. Zuck, and C. Poellabauer, “Workload-aware dual-speed dynamic voltage scaling,” in Proc. of IEEE RTCSA, 2006. [7] N. Balasubramanian, A. Balasubramanian, and A. Venkataramani, “Energy consumption in mobile phones: a measurement study and implications for network applications,” in Proc. of ACM SIGCOMM, 2009. [8] M. Dong and L. Zhong, “Self-constructive high-rate system energy modeling for battery-powered mobile systems,” in Proc. of ACM MobiSys, 2011. [9] F. Xu, Y. Liu, Q. Li, and Y. Zhang, “V-edge: Fast self-constructive power modeling of smartphones based on battery voltage dynamics.” in Proc. of NSDI, 2013. [10] X. Chen, Y. Chen, Z. Ma, and F. C. Fernandes, “How is energy consumed in smartphone display applications?” in Proc. of ACM HotMobile, 2013. [11] R. LiKamWa, B. Priyantha, M. Philipose, L. Zhong, and P. Bahl, “Energy characterization and optimization of image sensing toward continuous mobile vision,” in Proc. of ACM MobiSys, 2013. [12] H. Han, J. Yu, H. Zhu, Y. Chen, J. Yang, G. Xue, Y. Zhu, and M. Li, “E 3: energy-efficient engine for frame rate adaptation on smartphones,” in Proc. of ACM Sensys, 2013. [13] T. Hao, G. Xing, and G. Zhou, “isleep: unobtrusive sleep quality monitoring using smartphones,” in Proc. of ACM Sensys, 2013. [14] J. R. Kwapisz, G. M. Weiss, and S. A. Moore, “Activity recognition using cell phone accelerometers,” SIGKDD, vol. 12, no. 2, pp. 74–82, 2011. [15] Y.-S. Lee and S.-B. Cho, “Activity recognition using hierarchical hidden markov models on a smartphone with 3d accelerometer,” in Proc. of Springer HAIS, 2011. [16] M. Shahzad, A. X. Liu, and A. Samuel, “Secure unlocking of mobile touch screen devices by simple gestures: You can see it but you can not do it,” in Proc.of ACM MobiCom, 2013. [17] Z. Chen, M. Lin, F. Chen, N. D. Lane, G. Cardone, R. Wang, T. Li, Y. Chen, T. Choudhury, and A. T. Campbell, “Unobtrusive sleep monitoring using smartphones,” in Proc. of IEEE PervasiveHealth, 2013. [18] D. A. Johnson and M. M. Trivedi, “Driving style recognition using a smartphone as a sensor platform,” in Proc. of IEEE ITSC, 2011. [19] C. Bo, L. Zhang, X.-Y. Li, Q. Huang, and Y. Wang, “Silentsense: silent user identification via touch and movement behavioral biometrics,” in Proc. of ACM MobiCom, 2013. 72