ENERGY EFFICIENT ALGORITHMS AND TECHNIQUES FOR WIRELESS MOBILE CLIENTS BHOJAN ANAND (Ph.D.), NUS A THESIS SUBMITTED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY DEPARTMENT OF COMPUTER SCIENCE SCHOOL OF COMPUTING NATIONAL UNIVERSITY OF SINGAPORE 2012 DECLARATION I hereby declare that the thesis is my original work and it has been written by me in its entirety I have duly acknowledged all the sources of information which have been used in the thesis This thesis has also not been submitted for any degree in any university previously ———————————– Bhojan Anand August 2012 iii iv ACKNOWLEDGEMENTS First of all I would like to thank my supervisor Professor A.L Ananda I appreciate his guidance and support not only my research but also my life After four years of his supervision, his impressive leadership becomes a big milestone in my life I am indebted to the coauthors of several papers included in this thesis The coauthors are Associate Professor Chan Mun Choon, Associate Professor Rajesh Krishna Balan, Associate Professor Ooi Wei Tsang, Associate Professor Chang Ee Chien, Mr Pravein Govindan Kannan, Mr Karthik Thirugnanam, Mrs Jeena Sebastien, Mr Le Thanh Long, Mr Pham Duc-Dung, Mr Soh Yu Ming and Mr Chong Lee Kee A special thanks to Associate Professor Ooi Wei Tsang and Associate Professor Roger Zimmerman for their productive comments for my thesis proposal and research I am also particularly grateful to Associate Professor Chan Mun Choon for his constructive comments and support throughout the research period I appreciate to all fellow Ph.D students and friends at the Communication and Internet Research Lab for making the supportive work environment My deepest gratitude goes to my wife Florence Anand and sons Mikhil Anand and Sashil Anand for their unflagging love and support throughout my life; this dissertation is simply impossible without them v TABLE OF CONTENTS Page DECLARATION iii ACKNOWLEDGEMENTS v SUMMARY xii LIST OF TABLES xv LIST OF FIGURES xvi LIST OF PUBLICATIONS xxix CHAPTER INTRODUCTION 1.1 1.2 1.3 Mobile Clients Space Mobile Applications Saving Energy 1.3.1 1.3.2 Why is saving energy in mobile clients important? Current Status & Challenges 1.3.2.1 1.3.2.2 1.4 1.5 Display Power Conservation 10 Wireless Interface Power Conservation 12 Thesis Contribution 13 Thesis Organisation 16 RELATED WORK 18 2.1 2.2 2.3 2.4 2.5 LCD Power Conservation 18 OLED Display Power Conservation 33 Network Interface Power Conservation 43 Processing Unit Power Conservation 61 Summary 63 vi DISPLAY POWER MANAGEMENT (LCD) 65 3.1 3.2 3.3 3.4 LCD Display Technology 65 Tone Mapping Technique & Its Advantages 68 Using the Gamma Function for Tone Mapping 70 System Design 73 3.4.1 3.4.2 Backlight Power Measurement 75 Gamma to Backlight Relationship 76 3.4.2.1 3.4.3 Measuring Image Quality 78 3.4.3.1 3.4.4 3.4.5 3.5 Objective Analysis of Gamma Thresholds 86 Run-time Algorithm 87 Selection of Games 88 Changing Backlight Level and Gamma 90 Evaluation Methodology 90 3.6.1 3.6.2 3.6.3 3.7 Objective 83 Methodology and Setup 83 Results 85 Implementation 88 3.5.1 3.5.2 3.6 Using the Image Quality Metrics 81 Computing Image Brightness 82 Human Calibration of Gamma Thresholds 83 3.4.5.1 3.4.5.2 3.4.5.3 3.4.6 3.4.7 Analysis of Gamma to Backlight Relationship 77 Power Measurement Testbed Setup 91 Power Measurement - Methodology 93 User Study - Methodology, Participants, and Setup 94 Evaluation Results 96 3.7.1 3.7.2 3.7.3 3.7.4 3.7.5 Baseline Measurements 96 Measured Analytical Results 99 User Study Results 100 Overall Result: System Works Very Well 100 Summary 102 vii DISPLAY POWER MANAGEMENT (OLED) 103 4.1 4.2 4.3 OLED Display Technology 104 Key Observations on OLED Displays 105 Power Optimisation for Webpages - Texts 110 4.3.1 4.3.2 4.3.3 4.4 Colour Harmony 111 Brand Colour & Brand Identity 113 Chromatic and Achromatic Contrast & Colour Mapping 120 Power Optimisation for Webpages - Images 124 4.4.1 4.4.2 Luminance Adaptive Colour Transformation 126 HVS based Colour Transformation Algorithm 131 4.4.2.1 4.4.3 4.5 4.6 Adapting to Other Contents 138 System Implementation 139 Evaluation Methodology 141 4.6.1 4.6.2 4.7 Algorithm Alternative 134 Quality Measurements - Objective Metrics 142 Quality Measurements - Subjective User Study 142 Evaluation Results 145 4.7.1 4.7.2 4.7.3 4.7.4 Evaluation Results of Colour Transformed Webpages 145 Evaluation results of HVS based Image Manipulation Algorithm 146 Overall Result (Combined) 150 Summary 151 POWER MANAGEMENT AT NETWORK INTERFACE LEVEL 152 5.1 Mobile Games and Game Maps 153 5.1.1 Binary Space Partitioning 154 5.1.1.1 5.1.1.2 5.1.2 5.2 Potentially Visible Set 155 Limitations 158 Quadtree and Octtree 158 Distance Based Approach 161 viii 5.2.1 5.2.2 5.2.3 5.3 Visibility Based Approach 169 5.3.1 5.3.2 5.4 Game State Estimation 161 Macro Power Management 162 Micro Power Management 166 Dynamic Lookahead 172 Determining Sleep Times and Intervals 175 3D Renderer’s View Based Approach 176 5.4.1 5.4.2 Visibility and Spatial Subdivision Scheme 177 Two-Level Scan Algorithms 178 5.4.2.1 5.4.2.2 5.5 5.6 5.7 Wireless Interface Control at Client Side 183 Algorithm Selection 184 Implementation 185 5.7.1 5.7.2 5.8 Network Characteristics of FPS and MMOG Games 185 Sleep Command 187 Evaluation Methodology 188 5.8.1 5.8.2 5.8.3 5.8.4 5.8.5 Evaluation Objectives 191 Defining a Quality & Power Metric 192 Small Scale User Study - Methodology 195 Experiments 197 Additional Evaluations - Visibility Based Approach 198 5.8.5.1 5.8.5.2 5.8.6 Using Traces for Repeatability 198 Running the Simulations 199 Additional Evaluations - Renderer’s View Based Approach 199 5.8.6.1 5.9 Macro Scanning Algorithm 179 Micro Scanning Algorithm 182 Using AMID for Error Control 200 Evaluation Results 203 5.9.1 Distance Based Approach 203 5.9.1.1 5.9.1.2 Sparse Environment - Low player Density 204 Dense Environment - High Player Density 205 ix 5.9.1.3 5.9.2 Visibility Based Approach 205 5.9.2.1 5.9.2.2 5.9.2.3 5.9.2.4 5.9.2.5 5.9.2.6 5.9.2.7 5.9.2.8 5.9.2.9 5.9.3 Contribution from Micro and Macro Power Management 205 Baseline - No Prediction 206 Experiments over Various Networks 208 Effect of Velocity 210 Effect of Different Sleep/Wakeup Intervals 212 Effect of player density 213 Benefit of Our Dynamic Algorithm 214 Real Power Measurements 215 Another Perspective of Power Savings 216 Impact of Errors on Perceived Quality 217 3D Renderer’s View Based Approach 218 5.9.3.1 5.9.3.2 5.9.3.3 5.9.3.4 5.9.3.5 Effects of Map Type 219 Effects of Energy Threshold 220 Effects of Player Density (number of Players) 221 Effects Error Controller on Optimising Algorithm Parameters 222 Effects Error Controller on Average Error Rate 224 5.10 Summary of all Results 226 CONCLUSIONS AND FUTURE WORK 227 6.1 Conclusions 227 6.1.1 6.1.2 6.1.3 6.2 Conserving LCD Energy 228 Conserving OLED Display Energy 229 Conserving Wireless Network Energy 229 Directions for the Future Research 230 BIBLIOGRAPHY 235 APPENDICES A GAMMA CALIBRATION (FOR LCD) - SURVEY FORM 259 B DISPLAY POWER MANAGEMENT (LCD) - SURVEY FORM 265 C DISPLAY POWER MANAGEMENT (OLED) - SURVEY APPLICATION AND FORM 273 x LIST OF TABLES Table Page 1.1 Battery Lifetime in Modern Smartphones 2.1 Power consumption (in Watts) for various modes of Intel (3x3 MIMO) and Atheros (2x2 MIMO) NICs 52 3.1 Demographics Statistics for the User Study 97 3.2 Baseline Power Consumption of the Laptop 98 3.3 Baseline Power Consumption of the HTC Magic 98 3.4 Power-Savings Measurements 99 4.1 Demographics Statistics for the User Study 143 5.1 Map Size of MMOG games 158 5.2 Power Characteristics of Different Interfaces 184 5.3 Selection of Algorithms 186 5.4 Power characteristic of the card used 189 5.5 Effects of Map Type - Experiment Variable Setup 219 xv LIST OF FIGURES Figure Page 1.1 Backlight level vs Power for HTC Magic Android Phone [1] 1.2 Component Power Consumption (HTC Magic) [1] 2.1 Structure of a Transmissive TFT LCD 19 2.2 Visibility of the Image in a Transmissive TFT in some Environment Luminance Condition [2] 20 2.3 Image and its Discrete Histogram 21 2.4 Luminance as a function of Backlight and Transmissivity 24 2.5 Visual Effects of Adjusting Brightness (b), Contrast (c), and Both (d) when the Backlight is Dimmed to 50% 25 2.6 Luminance as a function of Backlight and Transmissivity [3] 27 2.7 Luminance vs Perceived Brightness 28 2.8 Relation Between MSE and Backlight Level 32 2.9 Energy plot of the hues of categorical colours (from left to right: blue, red, purple, orange, green, yellow) Evergy (E) vs Lightness (L∗ ) [4] 35 2.10 Categorical colours of varying lightness sorted by increasing energy cost [4] 35 2.11 Tooth Dataset Coloured with Traditional Colours (Original) and Energy Efficient Colours [4] 36 2.12 Unstructured Transformed GUIs with Different Settings 38 2.13 Colour Transformed ESPN Webpage 39 xvi 2.14 Colour Quantizing an Image to N = 32 Colours 40 2.15 Colour Quantizing an Image to N = 512 Colours 42 2.16 Power Consumption of Images Generated using Different Algorithms 43 2.17 Power-constrained LMHE 44 2.18 802.11 State Transition 46 2.19 Client Listening Intervals in BSD 48 2.20 Drop Rate vs Energy Metric (Real player format at 512 Kbps) [5] 51 2.21 Split Communication for Multimedia Streaming 53 2.22 Cell-to-Notify Protocol 56 2.23 Suspension Period 60 2.24 Suspension Period 61 3.1 Transmissive LCD Displays 66 3.2 Reflective LCD Displays 67 3.3 The Effect of Gamma and Linear Transformations The Amount of Power Saved is the Same for Both Approaches 71 3.4 Global Contrast Loss vs Gamma 73 3.5 Global Contrast Change vs Image Brightness for γ = 74 3.6 Power vs Backlight level 75 3.7 Backlight level vs Gamma value 77 3.8 Effect of Gamma Increase and Compensation 78 3.9 Gamma and Backlight Adjusting Tool 84 3.10 User-Perceived Acceptable Gamma Levels 86 3.11 Contrast Loss vs Adaptive Gamma 87 xvii 3.12 Flowchart of our System 89 3.13 Testbed 92 3.14 Setup for Laptop Power Measurement (Overall) [Note: Lenovo Thinkpad W500 Laptop Adapter outputs 20V DC] 92 3.15 Results of the User Study for all Maps All Versions of the Game were Deemed Playable by the Participants 101 4.1 OLED Energy Consumption vs Screen Brightness 106 4.2 OLED Energy Consumption vs Gamma Value 107 4.3 AMOLED sub-pixels close-up 108 4.4 Energy Vs RGB Sub-Pixel Values 110 4.5 Colour Wheel in RGB Colour Space 112 4.6 Colours Wheel Types 114 4.7 Colours - Associated Words - Sample Logos 115 4.8 Webpages designed using Brand Colours available in their Logos 116 4.9 Logos are Used as Favicons 118 4.10 Colour Extraction and Ranking from NUS favicon 119 4.11 Colour Extraction and Ranking from INTEL favicon 119 4.12 Background vs Text Area (After Excluding Images) - Sample Webpages 123 4.13 Background vs Text Area 124 4.14 Change in Contrast vs Gamma 126 4.15 Effects of basic approaches on Image Contrast 127 4.16 Pixel power models of Goole Nexus One, Samsung Galaxy S and Nokia N85 OLED displays Axis X represents gamma-corrected linear RGB values [6] 129 xviii 4.17 Human Visual System Sensitivity 130 4.18 Local Contrast vs Brightness 131 4.19 An Image and its Contrast Map 132 4.20 Image Colour Transformation Process 134 4.21 Centralised Colour Transformation 140 4.22 Distributed Colour Transformation 141 4.23 Study on Readability and Colour Harmonicity 144 4.24 Study on Image Quality 145 4.25 Web Page Transformation with El-pincel 147 4.26 Web Page Transformation - User Study 148 4.27 Power consumption and quality measurements 149 4.28 Image Transformation - User Study 150 5.1 Renderer’s view of world 156 5.2 A Cluster Divided into Areas 157 5.3 MMOG Map Sizes - A Graphical Comparison 159 5.4 Level of Detail 160 5.5 AoI - Dynamic Hexagonal Tile Distance/Visibility 165 5.6 Distance Based Approach 168 5.7 Inter-player Visibility With Obstacles 169 5.8 Possible Locations After Three Time-steps 171 5.9 Basic Grid Weights (One Time-Step) 172 5.10 Scaling Grid Weights (Three Time-steps) 174 5.11 BFS follows the Path Distance of each Cluster 180 xix 5.12 Setup for Wireless Interface Power Measurement 189 5.13 PCMCIA Extender card (Accurite Technologies) 190 5.14 PCMCIA NEC PA-WL/54AG WiFi card used in the Experiments 190 5.15 Screen Short Measuring MPD 196 5.16 Error Control Loop with AMID 202 5.17 Energy saving in sparse environment 204 5.18 Energy saving in dense environment 205 5.19 Distance (Macro) and Angle (Micro) based algorithms composition 206 5.20 Results with no Prediction (q3dm1 Map) 207 5.21 Power Savings (3.5G Networks) 208 5.22 Power Savings (WiFi Networks) 209 5.24 Effect of Different sleep intervals 211 5.23 Effect of Velocity 211 5.25 Effect of Player Density 213 5.26 Dynamic Versus Static Algorithms 214 5.27 Actual Versus Simulation Results 215 5.28 User Study - Quality Loss Versus Alpha 217 5.29 MapType vs Energy Saved 220 5.30 Error Threshold vs Energy Saved 221 5.31 Number of Players vs Energy Saved 222 5.32 Fixed Sleep Time of Micro Scan varies over Time 223 5.33 Error Convergence Over the Game Play Time for ET=5% 223 xx 5.34 Error Convergence Over the Game Play Time for ET=3% 225 5.35 Error Convergence Over the Game Play Time for ET=1% 225 xxi NOMENCLATURE ACC Achromatic Contrast AER Average Error Rate AIMD Additive Increase Multiplicative Decrease AMOLED Active-matrix Organic Light Emitting Diode AP Access Point BL Backlight Level BS Base Station BSD Bounded Slowdown Protocol CAM Constantly Awake Mode CBCS Concurrent Brightness and Contrast Scaling CBL Content Brightness Level CBVLC Constant Backlight with Video Luminosity Compensation CC Chromatic Contrast CCFL Cold Cathode Fluorescent Lamps xxii CPSM Centralised Power Save Mode CTM Converse Tone Mapping CTS Clear To Send DAQ Data Acquisition Equipment DBA Distance Based Approach DBP Dynamic Beacon Period DCA Dual-Compensation Approach DCF Distributed Coordination Function DLS Dynamic Backlight Luminance Scaling DRA Dual Ring Algorithm DTM Dynamic Tone Mapping DVFS Dynamic Voltage and Frequency Scaling ESD Effective Sleep Duration ESPN Entertainment and Sports Programming Network FPS First Person Shooting Game FST Fixed Sleep Time GCL Global Contrast Change (Loss) GOP Group of Pictures xxiii GOS Group of Scenes GUI Graphical User Interface HAL Hardware Access Layer HD High Definition - Resolution (1280 x 720) HEBS Histogram Equalization for Backlight Scaling HSV Colour Space based on Hue, Saturation, Value or Lightness HVS Human Visual System JND Just Noticeable Deference LC Liquid-Crystal Cells LCD Liquid Crystal Display LED Light Emitting Diode Li-Ion Lithium-Ion Li-Po Lithium-Polymer LMHE Log-Modified Histogram Equalisation LMS Least Mean Squares LWM Low Water Mark Level MAC Media Access Control MIMO Multiple Input Multiple Output xxiv MMOG Massively Multiplayer Online Game MPD Vertically Integrated Approach MPEG Moving Picture Experts Group MPSM Mobility Aware Power Saving Mode MSE Mean Square Error MSSIM Mean Structural SIMilarity Index NAPman Network-Assisted Power Management NI National Instruments Ni-Cd Nickel-Cadmium Ni-MH Nickel-Metal Hydride OIA Open Innovation Approach OLED Organic Light Emitting Diode OS Operating System PB Perceived Brightness PCI Peripheral Component Interconnect PDVS Practical Voltage Scaling PID Proportional Integral Derivative PLRD Programmable LCD Reference Driver xxv PM Power Model based Colour mapping approach PMOLED Passive-matrix Organic Light Emitting Diode PS Poll Power Save Poll PSD Potential Sleep Duration PSM Power Saving Mode PSNR Peak Signal to Noise Ratio PVS Potentially Visible Sets Q3DM1 Quake III Death Match Q3DM7 Quake III Death Match QABS Quality Adaptive Backlight Scaling QoS Quality of Service RBA Renderer’s View Based Approach RC Resource Controller RGB Colour Space based on tristimulus colours Red, Green and Blue RTCP Real Time Control Protocol RTP Real Time Protocol RTPS Real Time and Power Saving RTS Request To Send xxvi RTT Round-Trip Time SAMOLED Super Active-matrix Organic Light Emitting Diode SC Signal Conditioning Equipment SPSM Smart Power Saving Mode SRA Single Ring Algorithm SSIM Structural Similarity Index STPM Self- Tuning Power Management TCP Transmission Control Protocol TFRC TCP-Friendly Rate Control TFT Thin-Film Transistor TIM Traffic Indication Map UAPSD Unscheduled Automatic Power Save Delivery URL Uniform Resource Locator VBA Visibility Based Approach VIA Vertically Integrated Approach VoIP Voice over Internet Protocol W3C World-Wide Web Consortium WLAN Wireless Local Area Networks xxvii WNIC Wireless Network Interface Card WoW World of Warcraft game WWW World Wide Web xxviii LIST OF PUBLICATIONS Bhojan Anand, Lee Kee Chong, Ee-Chien Chang, Mun Choon Chan, Akkihebbal L Ananda, Wei Tsang Ooi, El-pincel - A Painter Cloud Service for Greener Web Pages,, Proceeding of ACM MM 2012, Nara, Japan, Oct 2012 Karthik Thirugnanam, Bhojan Anand, Jeena Sebastian, Pravein Govindan Kannan, Akkihebbal L Ananda, Rajesh Krishna Balan, and Mun Choon Chan, Dynamic Lookahead Mechanism for Conserving Power in Multi-Player Mobile Games,, Proceeding of IEEE INFOCOM 2012, Orlando, Florida, Mar 2012 Bhojan Anand, Akkihebbal L Ananda, Mun Choon Chan and Rajesh Krishna Balan, ”ARIVU: Making Networked Mobile Games Green - A Scalable Power-Aware Middleware”, MOBILE NETWORKS AND APPLICATIONS, Springer Netherlands,(DOI: 10.1007/s11036-011-03128, URL:http://dx.doi.org/10.1007/s11036-011-0312-8), Feb 2012 Bhojan Anand, Karthik Thirugnanam, Jeena Sebastien, Pravein Govindan Kannan, Akkihebbal L Ananda, Mun Choon Chan and Rajesh Krishna Balan, ”Adaptive Display Power Management for Mobile Games”, Proceeedings of ACM MobiSys 2011, Washington DC, June 2011 (abstract paper) Bhojan Anand, Pravein Govindan Kannan, Akhihebbal L Ananda, Mun Choon Chan, ”El-pincel - A Painter Cloud Service for Greener Web Pages”, Proceeedings of ACM MobiSys 2011, Washington DC, June 2011 Bhojan Anand, Soh Yu Ming, Akkihebbal L Ananda, Mun Choon Chan, and Rajesh Krishna Balan, ”PGTP: Power Aware Game Transport Protocol for Multi-Player Mobile Games”, IEEE International Conference on Communications and Signal Processing (ICCSP), National Institute of Technology, Calicut, India, Feb 2011 Bhojan Anand, Karthik Thirugnanam, Le Thanh Long, Duc-Dung Pham, Akkihebbal L Ananda, Rajesh Krishna Balan, and Mun Choon Chan, ”ARIVU: Power-Aware Middleware for Multiplayer Mobile Games”, IEEE/ACM Netgames, Teipei, Taiwan, Nov 2010 Paper - Bo Han, Ahmad Rahmati, Bhojan Anand, ”Report of HotMobile 2010,” IEEE Pervasive Computing Journel, pp 93-96, July-September, 2010 (abstract paper, doctrol consortium) Bhojan Anand ”Konva: Power and Network Aware Framework and Protocols for Multiplayer Mobile Games”, Proceedings of The Eleventh ACM Workshop on Mobile Computing Systems & Applications (HotMobile 2010), Feburary 22, Annapolis, MD, USA Bhojan Anand, Akkihebbal L.Ananda, Mun Choon Chan, Le Thanh Long and Rajesh Krishna Balan, ”Game Action Based Power Management for Multiplayer Online Game”, Proceedings of ACM SIGCOMM Workshop on Networking, Systems, Applications on Mobile Handhelds (MobiHeld 2009), August 17, Barcelona, Spain xxix ... Chan and Rajesh Krishna Balan, ”ARIVU: Making Networked Mobile Games Green - A Scalable Power-Aware Middleware”, MOBILE NETWORKS AND APPLICATIONS, Springer Netherlands,(DOI: 10. 100 7/s1 103 6 -01 1 -03 128,... TECHNIQUES FOR WIRELESS MOBILE CLIENTS BHOJAN ANAND (Ph.D.), NATIONAL UNIVERSITY OF SINGAPORE Directed by: Professor Dr A.L Ananda In this thesis we present a suite of algorithms and techniques for. .. URL:http://dx.doi.org/ 10. 100 7/s1 103 6 -01 1 -03 12-8), Feb 201 2 Bhojan Anand, Karthik Thirugnanam, Jeena Sebastien, Pravein Govindan Kannan, Akkihebbal L Ananda, Mun Choon Chan and Rajesh Krishna Balan,