智慧型手機的功能越來越多元，便利的功能使耗電量增加，電源管理成為智慧型手機的重要議題。許多研究嘗試在使用者體驗與能源 節省間取得平衡。使用者體驗為使用者對於服務的主觀感受，受到服務 品質、使用者個人期待、使用者所在情境的影響。近年相關的研究多專 注於服務品質與使用者體驗的關聯，但未探討使用者情境對於使用者 體驗的影響，為此我們招募了八位受測者進行為期兩週的實驗，實驗結 果顯示使用者在不同的情境下，對於系統效能的要求有極大的差異。
在這篇論文中，我們提出了一個以使用者為中心的情境感知 CPU- GPU DVFS機制，並根據使用者當下情境決定如何調整系統資源。我們 也設計了一些評估實驗來做驗證。實驗結果顯示，與 Android 手機的 預設 DVFS 機制相比，我們所提倡的 DVFS 機制最多可節省 25%的能源消 耗。

With the increasing functionalities in smartphones, smartphones become an important part of our daily lives. Users can communicate, navigate, and check the updates of social media with smartphones anywhere at any time. However, due to the limit battery capacity, the power management has become a critical issue to the battery-driven smartphones.
A number of existing works strike to balance the user’s quality of experience(QoE) and the energy saving. However, most of them are unaware of the impact of the user’s context to the user’s QoE, which cause the over- or under-provisioning hardware. To observe the correlation between the user’s contexts and the user’s QoE, we conducted an experiment with 8 subjects for two weeks. The results reveal that the demand for users’ QoE varies in different contexts, which inspire us to scale the system performance based on context information to optimize energy saving. In this thesis, we propose a governor which optimize the energy saving while maintaining the user’s QoE simultaneously in the different contexts. The evaluation shows the proposed governor can save power consumption up to 25% compared to the default governor while keeping the users felt satisfied with the quality of service.

1 Introduction 1
2 Related Work 5
2.1 User-centric power management 5
2.2 User context 6
3 System Architecture 7
3.1 Overview 7
3.2 Stage detection 8
3.2.1 Response stage 8
3.2.2 Waiting stage .9
3.3 Context-aware CPU-GPUcontrollingservice 14
3.3.1 Usercontextcollection 14
3.3.2 Usercontextclassification 16
3.3.3 The correlation between the user context and the user QoE 18
3.3.4 Context-basedFDRcontrollingservice 21
4 Evaluation 23
4.1 Experiment setup 23
4.2 Power consumption compared with the different governor in different
contexts 24
4.3 User satisfaction with the compared governors in the different context 25
4.4 User-centered context-aware CPU-GPU governor overhead 28
5 Discussion 30
5.1 User study 30
5.2 Limitation 32
6 Conclusion 34

[1] https://play.google.com/store/apps/details?id=com.facebook. katana.
[2] https://play.google.com/store/apps/details?id=com.facebook.orca.
[3] https://play.google.com/store/apps/details?id=mong.moptt.
[4] https://play.google.com/store/apps/details?id=com.sparkslab. dcardreader.
[5] https://edition.cnn.com/2013/03/28/tech/mobile/ survey-phones-facebook/index.html.
[6] https://developer.android.com/reference/android/ accessibilityservice/AccessibilityService.
[7] https://developer.android.com/reference/android/view/View. OnTouchListener.
[8] https://play.google.com/store/apps/details?id=com.duapps. recorder.
[9] http://docs.opencv.org/2.4.13/.
[10] https://developer.android.com/guide/.
[11] https://developer.android.com/studio/command-line/adb.
[12] https://developer.android.com/studio/command-line/dumpsys.
[13] https://www.msoon.com/.
[14] https://play.google.com/store/apps/details?id=com.quicinc.trepn.
[15] Youngho Ahn and Ki-Seok Chung. User-centric power management for embedded cpus using cpu bandwidth control. IEEE Transactions on Mobile Comput- ing, 15(9):2388–2397, 2016.
[16] Kjell Brunnstr ̈om, Sergio Ariel Beker, Katrien De Moor, Ann Dooms, Sebastian Egger, Marie-Neige Garcia, Tobias Hossfeld, Satu Jumisko-Pyykko ̈, Christian Keimel, Mohamed-Chaker Larabi, et al. Qualinet white paper on definitions of quality of experience. 2013.
[17] Wei-Ming Chen, Sheng-Wei Cheng, Pi-Cheng Hsiu, and Tei-Wei Kuo. A user- centric cpu-gpu governing framework for 3d games on mobile devices. In Computer-Aided Design (ICCAD), 2015 IEEE/ACM International Conference on, pages 224–231. IEEE, 2015.
[18] Z. Cheng, X. Li, B. Sun, J. Song, C. Wang, and X. Zhou. Behavior-aware integrated cpu-gpu power management for mobile games. In 2016 IEEE 24th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS), pages 439–444, Sept 2016.
[19] P. Chuang, Y. Chen, and P. Huang. An adaptive on-line cpu-gpu governor for games on mobile devices. In 2017 22nd Asia and South Pacific Design Automation Conference (ASP-DAC), pages 653–658, Jan 2017.
[20] Anind K Dey. Understanding and using context. Personal and ubiquitous computing, 5(1):4–7, 2001.
[21] B. Dietrich, D. Goswami, S. Chakraborty, A. Guha, and M. Gries. Time series characterization of gaming workload for runtime power management. IEEE Transactions on Computers, 64(1):260–273, Jan 2015.
[22] Begum Egilmez, Matthew Schuchhardt, Gokhan Memik, Raid Ayoub, Niranjan Soundararajan, and Michael Kishinevsky. User-aware frame rate management in android smartphones. ACM Trans. Embed. Comput. Syst., 16(5s):131:1– 131:17, September 2017.
[23] Denzil Ferreira, Jorge Goncalves, Vassilis Kostakos, Louise Barkhuus, and Anind K Dey. Contextual experience sampling of mobile application micro- usage. In Proceedings of the 16th international conference on Human-computer interaction with mobile devices & services, pages 91–100. ACM, 2014.
[24] M. Fiedler, T. Hossfeld, and P. Tran-Gia. A generic quantitative relationship between quality of experience and quality of service. IEEE Network, 24(2):36– 41, March 2010.
[25] Kuan-Ting Ho, Chung-Ta King, Bhaskar Das, and Yung-Ju Chang. Characterizing display qos based on frame dropping for power management of interactive applications on smartphones. In Design, Automation & Test in Europe Conference & Exhibition (DATE), 2018, pages 873–876. IEEE, 2018.
[26] Mohammad Ashraful Hoque, Matti Siekkinen, and Jukka K. Nurminen. Using crowd-sourced viewing statistics to save energy in wireless video streaming. In Proceedings of the 19th Annual International Conference on Mobile Computing & Networking, MobiCom ’13, pages 377–388, New York, NY, USA, 2013. ACM.
[27] Mohammad Ashraful Hoque, Matti Siekkinen, Jukka K Nurminen, Mika Aalto, and Sasu Tarkoma. Mobile multimedia streaming techniques: Qoe and energy saving perspective. Pervasive and Mobile Computing, 16:96–114, 2015.
[28] Zhexue Huang. Clustering large data sets with mixed numeric and categorical values. In Proceedings of the 1st pacificasia conference on knowledge discovery and data mining,(PAKDD), pages 21–34. Singapore, 1997.
[29] Zhexue Huang. Extensions to the k-means algorithm for clustering large data sets with categorical values. Data mining and knowledge discovery, 2(3):283– 304, 1998.
[30] Chanyou Hwang, Saumay Pushp, Changyoung Koh, Jungpil Yoon, Yunxin Liu, Seungpyo Choi, and Junehwa Song. Raven: Perception-aware optimization of power consumption for mobile games. In Proceedings of the 23rd Annual International Conference on Mobile Computing and Networking, MobiCom ’17, pages 422–434, New York, NY, USA, 2017. ACM.
[31] Dongwon Kim, Nohyun Jung, and Hojung Cha. Content-centric display energy management for mobile devices. In 2014 51st ACM/EDAC/IEEE Design Automation Conference (DAC), pages 1–6, June 2014.
[32] Xianfeng Li, Guikang Chen, and Wen Wen. Energy-efficient execution for repetitive app usages on big. little architectures. In Design Automation Conference (DAC), 2017 54th ACM/EDAC/IEEE, pages 1–6. IEEE, 2017.
[33] Xin Li, Mian Dong, Zhan Ma, and Felix C.A. Fernandes. Greentube: Power optimization for mobile videostreaming via dynamic cache management. In Proceedings of the 20th ACM International Conference on Multimedia, MM ’12, pages 279–288, New York, NY, USA, 2012. ACM.
[34] James MacQueen et al. Some methods for classification and analysis of multivariate observations. In Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, volume 1, pages 281–297. Oakland, CA, USA, 1967.
[35] Livingstone Margaret and Huble David. Vision and art: The biology of seeing. New York, 2002.
[36] Anuj Pathania, Qing Jiao, Alok Prakash, and Tulika Mitra. Integrated cpu- gpu power management for 3d mobile games. In Design Automation Conference (DAC), 2014 51st ACM/EDAC/IEEE, pages 1–6. IEEE, 2014.
[37] Ulrich Reiter, Kjell Brunnstro ̈m, Katrien De Moor, Mohamed-Chaker Larabi, Manuela Pereira, Antonio Pinheiro, Junyong You, and Andrej Zgank. Factors influencing quality of experience. In Quality of experience, pages 55–72. Springer, 2014.
[38] Choonsung Shin, Jin-Hyuk Hong, and Anind K Dey. Understanding and prediction of mobile application usage for smartphones. In Proceedings of the 2012 ACM Conference on Ubiquitous Computing, pages 173–182. ACM, 2012.
[39] Wook Song, Nosub Sung, Byung-Gon Chun, and Jihong Kim. Reducing energy consumption of smartphones using user-perceived response time analysis. In Proceedings of the 15th Workshop on Mobile Computing Systems and Applications, HotMobile ’14, pages 20:1–20:6, New York, NY, USA, 2014. ACM.
[40] Moshe Unger, Ariel Bar, Bracha Shapira, Lior Rokach, and Ehud Gudes. Contexto: lessons learned from mobile context inference. In Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing: Adjunct Publication, pages 175–178. ACM, 2014.
[41] Katarzyna Wac, Selim Ickin, Jin-Hyuk Hong, Lucjan Janowski, Markus Fiedler, and Anind K. Dey. Studying the experience of mobile applications used in different contexts of daily life. In Proceedings of the First ACM SIGCOMM Workshop on Measurements Up the Stack, W-MUST ’11, pages 7–12, New York, NY, USA, 2011. ACM.
[42] Zhou Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing, 13(4):600–612, April 2004.
[43] Ye Xu, Mu Lin, Hong Lu, Giuseppe Cardone, Nicholas Lane, Zhenyu Chen, Andrew Campbell, and Tanzeem Choudhury. Preference, context and communities: a multi-faceted approach to predicting smartphone app usage patterns. In Proceedings of the 2013 International Symposium on Wearable Computers, pages 69–76. ACM, 2013.
[44] Lei Yang, Robert P Dick, Gokhan Memik, and Peter Dinda. Happe: Human and application-driven frequency scaling for processor power efficiency. IEEE transactions on mobile computing, 12(8):1546–1557, 2013.
[45] Semir Zeki. A vision of the brain. Blackwell Scientific Publ., 1993.