1st Edition

Generic and Energy-Efficient Context-Aware Mobile Sensing

By Ozgur Yurur, Chi Harold Liu Copyright 2015
    222 Pages
    by CRC Press

    222 Pages 41 B/W Illustrations
    by CRC Press

    Elaborating on the concept of context awareness, this book presents up-to-date research and novel framework designs for context-aware mobile sensing. Generic and Energy-Efficient Context-Aware Mobile Sensing proposes novel context-inferring algorithms and generic framework designs that can help readers enhance existing tradeoffs in mobile sensing, especially between accuracy and power consumption.

    The book presents solutions that emphasize must-have system characteristics such as energy efficiency, accuracy, robustness, adaptability, time-invariance, and optimal sensor sensing. Numerous application examples guide readers from fundamental concepts to the implementation of context-aware-related algorithms and frameworks.

    Covering theory and practical strategies for context awareness in mobile sensing, the book will help readers develop the modeling and analysis skills required to build futuristic context-aware framework designs for resource-constrained platforms.

    • Includes best practices for designing and implementing practical context-aware frameworks in ubiquitous/mobile sensing
    • Proposes a lightweight online classification method to detect user-centric postural actions
    • Examines mobile device-based battery modeling under the scope of battery nonlinearities with respect to variant loads
    • Unveils a novel discrete time inhomogeneous hidden semi-Markov model (DT-IHS-MM)-based generic framework to achieve a better realization of HAR-based mobile context awareness

    Supplying theory and equation derivations for all the concepts discussed, the book includes design tips for the implementation of smartphone programming as well as pointers on how to make the best use of MATLAB® for the presentation of performance analysis. Coverage includes lightweight, online, and unsupervised pattern recognition methods; adaptive, time-variant, and optimal sensory sampling strategies; and energy-efficient, robust, and inhomogeneous context-aware framework designs.

    Researchers will learn the latest modeling and analysis research on mobile sensing. Students will gain access to accessible reference material on mobile sensing theory and practice. Engineers will gain authoritative insights into cutting-edge system designs.

    Context Awareness for Mobile Sensing
    Introduction
    Context Awareness Essentials
         Contextual Information
         Context Representation
         ContextModeling
         Context-Aware Middleware
         Context Inference
         Context-Aware Framework Designs
    Context-Aware Applications
         Health Care andWell-Being Based
         Human Activity Recognition Based
         Transportation and Location Based
         Social Networking Based
         Environmental Based
    Challenges and Future Trends
         Energy Awareness
         Adaptive and Opportunistic Sensory Sampling
         Modeling the Smart Device Battery Behavior for Energy Optimizations 
         Data Calibration and Robustness 
         Efficient Context Inference Algorithms
         Generic Context-Aware Framework Designs
         Standard Context-Aware Middleware Solutions
         Mobile Cloud Computing 
         Security, Privacy, and Trust

    Context Inference: Posture Detection
    Discussions
    Proposed Classification Method
    Standalone Mode
    Assisting Mode
         Feature Extraction
         Pattern Recognition–Based Classification
              Gaussian Mixture Model
              k-Nearest Neighbors Search
              Linear Discriminant Analysis
         Online Processing: Dynamic Training 
         Statistical Tool–Based Classification
    Performance Evaluation

    Context-Aware Framework: A Basic Design
    Discussions
    Proposed Framework
         Preliminaries
         User State Representation
         System Adaptability 
              Time-Variant User State Transition Matrix
              Time-Variant Observation Emission Matrix
              Update on System Parameters
              Entropy Rate
              Scaling Problem
    Simulations
         Preparations
         Applied Process
         Power Consumption Model
         Accuracy Model
         Parameter Setups
         Results and Discussions
    Validation by a Smartphone Application
         Observation Analysis
              Construction of Observation Emission Matrix
         Applied Process
         Performance Evaluation

    Energy Efficiency in Physical Hardware
    Discussions
    Battery Modeling
    Modeling of Energy Consumption by Sensors
         Preliminaries
         Modeling of Sensory Operations
    Validation by a Smartphone Application
    Sensor Management
         Battery Case
         Sensor Utilization Case
    Performance Analysis
         Method I (MI)
         Method II (MII)
         Method III (MIII)

    Context-Aware Framework: A Complex Design
    Proposed Framework
    Context Inference Module
         Inhomogeneous Statistical Machine
              Basic Definitions and Inhomogeneity
              Underlying Process
              User State Representation
              Time-Variant User State TransitionMatrix
              Adaptive Observation Emission Matrix
         Accuracy Notifier and Definition of Actions
    Sensor Management Module
         Sensor Utilization
         Trade-Off Analysis
         Intuitive Solutions
              Method I (MI)
              Method II (MII)
              Method III (MIII)
         Constrained Markov Decision Process–Based Solution
         Partially Observable Markov Decision
    Process–Based Solution
              Myopic Strategy and Sufficient Statistics
    Performance Evaluation

    Probabilistic Context Modeling
    Construction of Hidden Markov Models
         General Model
         Parallel HMMs
         Factorial HMMs
         Coupled/Joint HMMs
         Observation Decomposed/Multiple Observation HMMs
         Hierarchical HMMs
         Dynamic Bayesian Networks
    Evaluation
    Inference
         Learning: Forward–Backward Procedure
         Extended Forward–Backward Procedure
    Model for Multiple Sensors Use

    Appendix
    References
    Index

    Biography

    Ozgur Yurur received a double major from the Department of Electronics Engineering and the Department of Computer Engineering at Gebze Institute of Technology, Kocaeli, Turkey, in 2008, and MSEE and PhD from the Department of Electrical Engineering at the University of South Florida (USF), Tampa, Florida, in 2010 and 2013, respectively. He is currently with RF Micro Devices, responsible for the research and design of new test development strategies and also for the implementation of hardware, software, and firmware solutions for 2G, 3G, 4G, and wireless-based company products. In addition, Dr. Yurur conducts research in the field of mobile sensing. His research area covers ubiquitous sensing, mobile computing, machine learning, and energy-efficient optimal sensing policies in wireless networks. The main focus of his research is on developing and implementing accurate, energy-efficient, predictive, robust, and optimal context-aware algorithms and framework designs on sensor-enabled mobile devices.

    Chi Harold Liu is a full professor at the School of Software, Beijing Institute of Technology, China. He is also the deputy director of IBM Mainframe Excellence Center (Beijing), director of IBM Big Data Technology Center, and director of National Laboratory of Data Intelligence for China Light Industry. He holds a PhD from Imperial College, United Kingdom, and a BEng from Tsinghua University, China. Before moving to academia, he joined IBM Research, China, as a staff researcher and project manager and was previously a postdoctoral researcher at Deutsche Telekom Laboratories, Germany, and a visiting scholar at IBM T. J. Watson Research Center, Armonk, New York. Dr. Liu’s current research interests include the Internet of Things (IoT), big data analytics, mobile computing, and wireless ad hoc, sensor, and mesh networks. He received the IBM First Plateau Invention Achievement Award in 2012 and an IBM First Patent Application Award in 2011. He was interviewed by EEWeb.com as the featured engineer in 2011.

    Dr. Liu has published more than 50 prestigious conference and journal papers and owns more than 10 EU, U.S., and China patents. He serves as the editor for KSII Transactions on Internet and Information Systems and was book author or editor of three books published by CRC Press. He has served as the general chair of the IEEE SECON’13 workshop on IoT Networking and Control, the IEEEWCNC’12 workshop on IoT Enabling Technologies, and the ACM UbiComp’11Workshop on Networking and Object Memories for IoT. He has also served as a consultant for Bain & Company and KPMG, United States; and as a peer reviewer for Qatar National Research Foundation and the National Science Foundation in China. He is a member of the IEEE and the ACM.