Energy awareness based on individual user activity on the mobile phone

Number of pages: 110 File Format: word File Code: 31034
Year: 2014 University Degree: Master's degree Category: Computer Engineering
  • Part of the Content
  • Contents & Resources
  • Summary of Energy awareness based on individual user activity on the mobile phone

    Dissertation for Master's Degree in Computer Science (M.Sc)

    Software Orientation

    Abstract

    Smartphones and other mobile devices are becoming an ideal platform for continuous measurement of user activities by a large number of embedded sensors. In this way, energy consumption has become the most important challenge of mobile devices. Detecting individual activities on smart phones still seems to be a challenge due to resource limitations such as battery life, computational workload. By considering the user's activity and its management, it is possible to bring low energy consumption to mobile phones and other mobile devices, which requires a complete and perfect programming to detect the activities and adjust the energy consumption of the device according to their use in different times and places. For this purpose, we have considered a smart energy management system based on user activity for smartphones using the Android operating system. Finally, this program, which seems necessary for both mobile device developers and users, aims to save 15% of energy in mobile phones. Keywords: human activity detection, motion sensors, energy, Android.

    Energy efficient refers to doing a task efficiently and optimally, either in the form of supporting services or in the form of processing a task, with the least use of available energy resources [1] [2]. The importance of the topic in the cause of the discussion of efficient energy in mobile computing devices is that recently the Internet has been directly affected by mobile devices (especially mobile phones in recent years). Cloud computing, a technology whose perspective is parallel processing, can be considered a clear example of the need for energy in mobile computing devices. In general, mobile computing is a chain technology whose applications are expanding day by day and one should look for a way to save the energy of the devices in question. With technologies such as 4G, 3G, CDMA, Wi-Fi and WiMax, mobility support is rapidly developing in parallel with the Internet to provide mobility-based services. As a result, mobile devices require efficient energy or sufficient power to support the user to access services for a long time. In general:

    Providing service to any user at any place and at any time means supporting the properties of accessibility/portability while moving.

    1-2-Limitations of the research

    Considering that the mobile phone device based on human-centered measurement can produce information about the user's location based on the information collected from the sensors, in order to recognize and clean each user's location after classifying his activities, there is a continuous need to interact with all the sensors in a device. We have a mobile phone. However, the continuous use of sensors will drain the battery of the mobile device. Therefore, it seems necessary to create a framework for using the sensor to accurately detect the user's positions with less energy consumption and manage the energy consumption by changing different profiles according to the user's needs. In this report, some strategies are proposed to correct these limitations.

    1-3-Importance of the subject

    Mobile devices have the most popularity and usage among communication devices, which encourages us to research to reduce energy consumption. Today, it is possible to detect user activity in mobile devices by embedded sensors, which can be used to manage energy in mobile devices by predicting user activity. In order to realize this goal, storing the characteristics of activities and classifying them and mapping them on the learning algorithm has been investigated. 1-4-Research Objective Our research is to introduce a method for detecting activity using the lever of predictable human behaviors to conserve energy by dynamically selecting sensors and disabling unnecessary expensive sensors and communications (in terms of battery energy consumption).

    1-4-Research Objective

    Our research is to introduce a method for detecting activity using the lever of predictable human behaviors to conserve energy by dynamically selecting sensors and disabling unnecessary expensive sensors and communications (in terms of battery energy consumption), which will greatly help in long-term energy conservation due to the limited energy in mobile phones.

    1-5-Research Method

    As It was mentioned in the previous discussions that the initial stage of the work is activity detection (activity detection) in the mobile device, which will be achieved by using sensors in accordance with Android programming. The output of the initial phase of the Android program in the simple SQLITE database

    Activity detection in Android

    Energy management in Android

    1-5-2- The reason for choosing Android programming for mobile activity detection

    Since the use of various electronic sensors in computer hardware became popular, a new spirit was breathed into the appearance and the way of using applications and entertainment programs. The use of these sensors was initially used due to the different needs of computer game devices and then to make the device easier to use. The Android operating system includes the ability to use these types of sensors in the program, and these sensors can be easily used to improve the quality of communication with the user [1]. Note that the programming environment used here is Eclipse and the programs will be written based on Android version 2.3 and above. This part of the research tries to examine the sensors in the simplest and fastest way and write a simple application program for it.  

    1-6- Research steps

    Overview of energy supply in the sensor

    Overview of the detection mechanism (detection) of user activity in computing devices

    Activity detection in Android

    Compilation of research history of activity detection in mobile devices

    Compilation of research history in energy awareness based on individual user activity in mobile devices

    Introduction of sensors Embedded in smartphones

    Dataset collection for this report

    Implementation of the researcher's idea

    1-7-Thesis structure

    In chapter 1, we had an overview of energy harvesting in the sensor from the environment, which can be effective in solving the big challenge of supporting mobility in mobile computing devices, which goes hand in hand with energy management to supply energy to mobile phones. In chapter 2, the history of general research in the field of detecting human activity by sensors, the history of research in detecting activity in mobile devices and the history of research in awareness of energy based on the individual activity of the user in mobile devices are presented, and chapter 3 will have an overview of the mechanism of detection (detection) of user activity in computing devices and the tools to achieve this goal, in chapter 4, we have mentioned data analysis, and in chapter 5, we have summarized the analysis and evaluation of the designed program, and finally, the final part of this chapter presents the idea of ??the researcher.

    Chapter Two

    Research history

    According to the requirements of the research, the research history in this report includes 4 parts, which we will deal with in order:

    Research history of harvesting energy from the environment to provide energy for mobile devices

    Research history in getting to know mobile sensors and smartphones

    Research history Activity detection in mobile devices

    Research history of energy awareness based on individual user activity in mobile devices

    2-1-Objective

    The activity detection approach based on execution threshold is able to distinguish between static and dynamic activities. Static activities refer to situations where the user is in a stationary state, such as standing, sitting, or lying down. Dynamic activities refer to activities that involve user movement. For example, walking, climbing stairs, and transitioning from sitting to standing [3]. The results obtained in the dataset collected by SmartLab [4] show that the accelerometer sensor alone can be used for accurate physical activity detection and the implementation strategy of activity classification algorithms in software environments.

  • Contents & References of Energy awareness based on individual user activity on the mobile phone

    List:

    Abstract. 1

    Chapter One: General Research

    1-1-Definition. 3

    1-2-Research limitation. 3

    1-3-The importance of the subject. 4

    1-4- Research objective. 4

    1-5-Research method. 4

    1-5-1- Steps to achieve activity detection in Android. 4

    1-5-2- The reason for choosing Android programming to detect mobile activity. 4

    1-6-steps of research. 5

    1-7-Thesis structure. 5

    Chapter Two: A review of research literature and research background

    2-1-Objective. 8

    2-2- Extracting energy from the environment to supply energy to mobile devices. 9

    2-2-1-Overview of energy harvesting in mobile systems. 9

    2-2-2-Principles of harvesting energy using human body heat. 11

    2-2-3- Using the human body as a heat source for mobile sensors. 12

    2-2-4-using TEGs in mobile devices. 14

    2-3-Research history in getting to know mobile phone sensors and smartphones. 15

    2-3-1-Sensors 15

    2-3-2-Smartphones. 18

    2-4-Research history of human activity detection by sensors 19

    2-4-1-Definition and classification of human activities. 20

    2-4-2-approaches to detect human activity. 24

    2-4-2-1-video-based approach. 24

    2-4-2-2-approach based on sensor environment. 24

    2-4-2-3-approach based on wearable sensors. 24

    2-4-3- approaches to modeling human activity. 25

    2-4-3-1- feature extraction approach. 25

    2-4-3-2-standard data and feature size reduction. 26

    2-4-4-Activity detection algorithms. 26

    2-4-4-1-clustering algorithms. 28

    2-4-4-2-K-nn clustering algorithm. 29

    2-4-4-3-ANN clustering algorithm. 29

    2-4-4-4-SVM clustering algorithm. 29

    2-4-4-5-Baysian clustering algorithm. 30

    2-4-4-6-Naïve Bayes clustering algorithm. 30

    2-4-4-7-Markov chain clustering algorithm. 30

    2-4-4-8-HMM clustering algorithm. 30

    2-4-4-9-Fuzzy Logic clustering algorithm. 31

    2-4-5-challenges 32

    2-4-5-1-complexity of activities 32

    2-4-5-2-number of activities 32

    2-4-5-3-type of activity. 32

    2-4-5-4-Educational data requirements. 33

    2-4-5-5-need for accuracy. 33

    2-4-5-6-long-term and high-level activity. 33

    2-4-6-sensor requirements. 34

    2-4-7-location of sensors 34

    2-4-8-real-time detection. 34

    2-4-9-pattern of human activity. 35

    2-4-10- A simple example of work on the detection of human activities. 37

    2-5- History of research on activity detection in mobile phones. 38

    2-6-Summary. 51

    The third chapter. 57

    Research tools. 57

    Mechanism for detecting (detecting) user activity in computing devices. 57

    The third chapter: Research implementation method

    3-1-Introduction. 58

    3-2- Introduction to Android programming. 58

    3-2-1-Problems. 59

    3-2-2-Applications 59

    3-2-3-A simple example of revealing human activities in the home environment. 60

    3-3- Display hierarchy. 61

    3-4-Film detection. 62

    3-5-sound events. 62

    3-6-Characteristics of shape and color. 62

    3-7-IE detection. 63

    3-8- Detection of GF and GE. 63

    3-9-Detecting human activities from behind obstacles using the animation of Doppler radar signals 63

    3-10-Detecting through RFID. 64

    3-11-Activity detection in Android. 65

    3-12-Determining user activity in mobile (achievements, challenges, recommendations). 65

    Chapter Four: Data Analysis

    4-1-Introduction. 71

    4-2-Data collection 71

    4-3-Data generation 72

    4-4-Program data collection. 74

    4-5-sensor management. 75

    4-6-Location management. 75

    4-7-File management. 76

    4-8-activity class. 76

    4-9-Steps of implementing the program. 77

    4-10- Feature extraction. 84

    4-11-Activity identification. 87

    Chapter Five: Conclusions and Suggestions

    5-1-Summary. 90

    5-2-Future work. 94

    List of sources. 96

    Appendix. 100

     

     

    Source:

     

    [1]     "glossary, "energy efficiency" " 12 August[2] D. Bertozzi, L. Benini, and B. Ricco', "Power Aware Network Interface Management for Streaming Multimedia," 2002. [3] D. Anguita, A. Ghio, L. Oneto, X. Parra, and J. L. Reyes-Ortiz, "Human Activity Recognition on Smartphones using a Multiclass Hardware-Friendly Support Vector Machine," 2012.

    [4] J. Raquel, "Smartphone Based Human Activity Prediction," 23rd July, 2013.

    [5] D. Anguita, A. Ghio, L. Oneto, X. Parra, and J. L. Reyes-Ortiz, "A Public Domain Dataset for Human Activity Recognition Using Smartphones," European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, April 24-26 2013. [6] V. Leonov, "Energy Harvesting for Self-Powered Wearable Devices," 2011. [7] V. Pop, J. v. d. Molengraft, F. Schnitzler, J. Penders, R. v. Schaijk, and R. Vullers, "POWER OPTIMIZATION FOR WIRELESS AUTONOMOUS TRANSDUCER SOLUTIONS," Proceedings of PowerMEMS 2008+ microEMS 2008, Sendai, Japan, November 9-12 2008.

    [8] J. D. Hardy and A. P. Gagge, "PHYSIOLOGICAL AND BEHAVIORAL TEMPERATURE REGULATION IN MEN SUPPRESSING AUSTRALIAN SUMMER BUSHFIRES WITH HANDS." TOOLS," 1970. [9] A. M. Khan, "Human Activity Recognition Using A Single Tri-axial Accelerometer," February, 2011 2011. [10] R. YU, "Analog, MEMS and Sensors enable our Mobile Devices into a SMART world," 2003. [11] L. Bao and S. S. Intille, "Activity Recognition from User-Annotated Acceleration Data," 2004.

    [12] M. JOSEFA and V. NADALES, "RECOGNITION OF HUMAN MOTION RELATED ACTIVITIES FROM SENSORS," 2010.

    [13] l. bao, "physical activity recognition from acceleration data under semi-naturalistic conditions," August 2003.

    [14] A. Bayat and M. Pomplun, "A Study on Human Activity Recognition Using Accelerometer Data from Smartphones," Procedia Computer Science, pp. 450 – 457, 2014.

    [15] A. BULLING, M. Planck, U. BLANKE, and B. SCHIELE, "A Tutorial on Human Activity Recognition Using Body-worn Inertial Sensors," p. 32 pages, (June 2013.

    [16] L. Atallah, B. Lo, R. King, and G.-Z. Yang, "Sensor Placement for Activity Detection using Wearable Accelerometers," 2010.

    [17] A. G. Wilde, "An Overview of Human Activity Detection Technologies for Pervasive Systems," 2010.

    [18]  ?. Yürür and W. Moreno, "Energy Efficient Sensor Management Strategies in Mobile Sensing," 2011.

    [19]   S. Iyer, R. Mayo, L. Luo, and P. Ranganathan, "Energy-Adaptive Display System Designs for Future Mobile Environments," Mobile Systems, Applications, and Services, April 23 2003.

    [20]   P. Pillai and K. G. Shin, "Real-Time Dynamic Voltage Scaling for Low-Power Embedded Operating Systems," 2001. [21] F. Martelli, "Wireless Sensor Networks," 2012. [22] R. N. Mayo and P. Ranganathan, "Energy Consumption in Mobile Devices: Why Future Systems Need Requirements-Aware Energy Scale-Down," 2005. [23] M. A. Sharaf, J. Beaver, A. Labrinidis, and P. K. Chrysanthis, "TiNA: A Scheme for Temporal CoherencyAware inNetwork Aggregation," September 19 2003.

    [24] S. N. Srirama, H. Flores, and C. Paniagua, "Zompopo: Mobile Calendar Prediction based on Human Activities Recognition using the Accelerometer and Cloud Services," 2011.

    [25]  J. A. Madni, R. R. Basava, and E. Chen, "Energy Efficient Localization Framework for Mobile Applications," The Fifth International Conference on Sensor Technologies and Applications, 2011.

    [26]   Z. Zhuang, K.-H. Kim, and J. P. Singh, "Improving Energy Efficiency of Location Sensing on Smartphones," 2010.

    [27] Y. Chon, E. Talipov, H. Shin, and H. Cha, "Mobility Prediction-based Smartphone Energy Optimization for Everyday Location Monitoring," 2011.

    [28] Jeongyeup, P. Joongheon, and K. R.

Energy awareness based on individual user activity on the mobile phone