Mobile and Ubiquitous Systems: Computing, Networking and Services: 19th EAI International Conference, MobiQuitous 2022 Pittsburgh, PA, USA, November 14–17, 2022 Proceedings

Description

This book constitutes the refereed post-conference proceedings of the 19th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, MobiQuitous 2022, which was held in Pittsburgh, November 14-17, 2022. The conference was held virtually due to the COVID-19 pandemic. The 26 full and 2 short papers were carefully reviewed and selected from 95 submissions and present discussions, They were organized in topical sections as follows: Internet of Things (IoT), Security and Privacy, Human-centric sensing, Drone applications and edge computing, Wireless networks, Mobile and human computer interactions, Poster and demo sessions, Technology for health Preface Organization Poster and Demo Sessions Audio Content and Crowdsourcing: A Subjective Quality Evaluation of Radio Programs Streamed Online Comparing Apples and Oranges: A Mobile User Experience Study of iOS and Android Consumer Devices Estimating the Area of Radio Signal Sources Activity in a Heterogeneous Sensor Network to Improve Battlefield Situational Awareness Forest Protection as a Context-Aware System Mobile Networks’ Analysis in Terms of QoS Performance Assessment Relationships Between QoS/QoE Metrics in Selected Mobile Network Measurement Campaign Contents Internet of Things (IoT) Devising a Vibration-Based Fault Detection System for Textile Machinery 1 Introduction 2 Background and Related Work 2.1 Vibration-Based Fault Detection 2.2 Photoelectric-Based Fault Detection 2.3 Fabric Defect-Based Machine Fault Detection 2.4 Machine Learning-Based Fault Detection 3 Proposed Methodology 3.1 System Design 3.2 Algorithm 4 Experimental Evaluation 4.1 Industrial Data Collection 4.2 Statistical Analysis 4.3 Findings 4.4 Fault Detection for Other Machinery 5 Real-Device Construction 6 Conclusion References Easing Construction of Smart Agriculture Applications Using Low Code Development Tools 1 Introduction 2 Motivation and Background 2.1 SOTA of LCDTs for Building SAAs 2.2 DisCoPar 3 DisCoPar-K 4 Validation and Discussion 4.1 Case Study 1: Monitoring Soil Moisture and Temperature 4.2 Case Study 2: Monitoring Humidity and Data Collection 4.3 Case Study 3: Tracking Application Data 4.4 Discussion 5 Conclusion and Future Work References R-MDP: A Game Theory Approach for Fault-Tolerant Data and Service Management in Crude Oil Pipelines Monitoring Systems 1 Introduction 2 Related Work 2.1 Placements and Communication Strategies in Fog Architectures 2.2 System Optimisation Using Game Theory 3 Data-Driven Resilient and Regionalised MDP (R-MDP) 3.1 Background and Problem Definition 3.2 Environment Setting and Model Definition 3.3 The First Stage: Accuracy Optimisation 3.4 The Second Stage: Optimising the Energy Consumption 4 Implementation and Results 4.1 Accuracy in Detection and Fault Tolerance 4.2 Optimised Accuracy and Energy Consumption 5 Conclusion and Future Work References UWB/IMU Fusion Localization Strategy Based on Continuity of Movement 1 Introduction 2 Related Work 3 Single Sensor Positioning Algorithm 3.1 Inertial Sensor Positioning Algorithm 3.2 Ultra-wideband Positioning Algorithm 4 Fusion Algorithm Based on UWB and IMU 5 Implementations and Evaluations 5.1 Experimental Setup 5.2 RWLS Algorithm Evaluations 5.3 ADKF Algorithm Evaluations 6 Conclusion References Security and Privacy Anonymous Yet Alike: A Privacy-Preserving DeepProfile Clustering for Mobile Usage Patterns 1 Introduction 2 Related Work 3 Method: DeepProfile Framework 3.1 DeepProfile Framework 3.2 Adaptive Time Segmentation Method 3.3 User Activity Partitioning 3.4 Deep Feature Extraction Model 3.5 Deep Feature Encoder 3.6 Spatio-Temporal-App Usage Clustering 4 Experiments and Results 4.1 Dataset 4.2 Adaptive Time Segmentation 4.3 Deep Feature Extraction Model and Clustering 4.4 Evaluation Results 4.5 Clustering Results from the Best Model 4.6 Cluster Transitions Analysis 5 Discussion 6 Conclusion and Future Work References Best-Practice-Based Framework for User-Centric Privacy-Preserving Solutions in Smart Home Environments 1 Introduction 2 The Framework for User-Centric Privacy-Preserving Approaches in Smart Homes 2.1 Proposed Framework 2.2 Integration of the Proposed Framework in User-Centric Privacy-Preserving Approach 3 Evaluation 3.1 Evaluation of Non-Functional Requirements 3.2 Evaluation of Functional Requirements 4 Discussion and Limitations 4.1 Discussion 4.2 Limitations 5 Related Work 6 Conclusions and Future Work References FedGroup: A Federated Learning Approach for Anomaly Detection in IoT Environments 1 Introduction 2 Literature Review 3 Methodology 3.1 IoT Datasets 3.2 Proposed Approach: FedGroup 3.3 Experiment and Analysis 4 Results 5 Conclusion References Privacy Protection Against Shoulder Surfing in Mobile Environments 1 Introduction 2 Related Work 3 Mobile Attack Scenario 4 User Interface as an Image (UIaaI) for Private Content Detection and Hiding 4.1 The Basic Idea 4.2 The Framework 4.3 Private Content Detection Network 4.4 An Image Augmentation Method for User-Defined Content Retraining 4.5 Private Content Blurring 4.6 Evaluation Results 5 Private Content Detection and Hiding for Web Applications 5.1 Background 5.2 Identifying Sensitive Content 5.3 Dynamic Detection of Sensitive Content 5.4 Blurring Sensitive Content via Applying CSS Styles 5.5 Evaluation 6 Conclusion and Future Work References Human-Centric Sensing A Shallow Convolution Network Based Contextual Attention for Human Activity Recognition 1 Introduction 2 Related Works 2.1 Feature Engineering Approaches 2.2 Deep Learning Approaches 3 Methdology 3.1 Rethinking Self-attention 3.2 Contextual Attention Mechanism 3.3 COA Based HAR Method 4 Experiment 4.1 Datasets 4.2 Experimental Platform 4.3 Values of Hyperparameters Used in the Baseline 4.4 Comparison with Other Methods 4.5 Ablation Studies 4.6 Visualizations 4.7 Discussion 5 Conclusion References Does Cycling Reveal Insights About You? Investigation of User and Environmental Characteristics During Cycling 1 Introduction 2 Related Work 3 Data Collection 3.1 Used Devices 3.2 Data Protection and Ethical Aspects 3.3 Recruitment 3.4 Smartphone and Smartwatch Placement 3.5 Scenario, Dimensions, and Parameter Variations 4 Data Processing 4.1 Preprocessing 4.2 Identification of Pedal Rotations 4.3 Feature Extraction 4.4 Classification 5 Results 5.1 Bike Type 5.2 Gear 5.3 Seat Height 5.4 Terrain 5.5 User Distinction 6 Limitations and Outlook 7 Conclusion References Feature Encoding by Location-Enhanced Word2Vec Embedding for Human Activity Recognition in Smart Homes 1 Introduction 2 Related Works 3 Proposed Methods 3.1 Data Pre-processing 3.2 Our Model 3.3 Evaluation Method 4 Experiment Results 4.1 Comparison Between Frequency Encoding and Word2Vec Embedding Encoding 4.2 Comparison Between None-Location Models and Location-Enhanced Models 4.3 Comparison Between Different SEW Size 4.4 Discussion 5 Conclusion and Future Work References HoneyBreath: An Ambush Tactic Against Wireless Breath Inference 1 Introduction 2 Preliminaries 2.1 Fresnel Zone 2.2 CSI-Based Breathing Rate Inference 3 Attack Model and Assumptions 4 Ambush Design 4.1 Overview 4.2 Planning Phase 4.3 Disturbance Manipulation 4.4 Breathing Rate Retrieval 4.5 From Point Ambush to Area Ambush 4.6 Security Analysis 5 Experimental Evaluation 5.1 Evaluation Setup 5.2 Breathing Rate Inference Attacks 5.3 Example Defenses 5.4 Overall Defense Impact 5.5 Two-User Scenario 5.6 Trap Area Evaluation 6 Related Work 7 Conclusion References LightSeg: An Online and Low-Latency Activity Segmentation Method for Wi-Fi Sensing 1 Introduction 2 Related Work 2.1 Threshold-Based Activity Segmentation 2.2 Deep Learning-Based Activity Segmentation 3 Design of LightSeg 3.1 Overview 3.2 Data Acquisition 3.3 Preprocessing 3.4 Activity Extraction 4 Evaluation 4.1 Experiment Settings 4.2 Performance of Activity Segmentation 4.3 Performance of Activity Classification 4.4 Evaluation of Computation Overhead 5 Conclusion References Revealing Mental Disorders Through Stylometric Features in Write-Ups 1 Introduction 2 Related Work 3 Our Proposed Methodology 3.1 Data Collection and Arrangement 3.2 Data Preparation 4 Evaluation Procedure 4.1 Classification Methods 4.2 Assessing Performances 5 Experimental Results 5.1 Mental Disorder Detection and Performance Measure 5.2 Class-Specific Performance Measure 5.3 Performance for Specific Pronouns 6 Discussion 7 Conclusion References RoomE and WatchBLoc: A BLE and Smartwatch IMU Algorithm and Dataset for Room-Level Localisation in Privacy-Preserving Environments 1 Introduction 2 The Dataset: WatchBLoc 2.1 Dataset Description 2.2 Challenges 3 System Architecture: RoomE 4 Room Detection 4.1 BLE Data Preprocessing 4.2 Algorithmic Details 4.3 Evaluation 5 Motion Detection 5.1 IMU Data Preprocessing 5.2 Algorithmic Details 5.3 Evaluation 6 Room Estimation 6.1 Model Definition 6.2 Inference 6.3 Parameter Learning 7 Related Work 8 Conclusion References Towards Cross Domain CSI Action Recognition Through One-Shot Bimodal Domain Adaptation 1 Introduction 2 Related Works 2.1 Model-Based Action Recognition 2.2 Learning-Based Cross Domain Action Recognition 2.3 Multi-modal Action Recognition 2.4 The Difference 3 Overview 4 Methodology 4.1 Data Preprocessing 4.2 Virtual Data Synthesis 4.3 Fusion of Amplitude and Phase 4.4 Action Classification 5 Evaluations 5.1 Experimental Setup and Datasets 5.2 Experimental Results of Activity Recognition 5.3 Experimental Results of Gesture Recognition 5.4 Comparison with Existing Works 5.5 Ablation Study 5.6 Parameter Study 6 Conclusions References Drone Applications and Edge Computing Sage: A Multiuser Cooperative Controller for Mobile Edge Systems 1 Introduction 2 Sage 2.1 Action Events and Configuration Settings 2.2 Action Event Processing 2.3 Source Nodes 2.4 Server Nodes 2.5 Dynamic Server Discovery 2.6 Adaptation 3 Use Case: Cooperative Games 4 Evaluation 4.1 Scalability 4.2 Adaptation 4.3 Computational Complexity 4.4 Churn 4.5 Smartphones as Mobile Servers 4.6 Energy Consumption 4.7 Granularity 5 Related Work 6 Conclusions and Future Work References Multi-MEC Cooperation Based VR Video Transmission and Cache Using K-Shortest Paths Optimization 1 Introduction 2 Related Work 3 System Architecture 3.1 Network Architecture 3.2 Transmission Process 4 Cache Strategy 4.1 Delay Analysis 4.2 Abstraction Method 4.3 Model Analysis 5 Experiment 5.1 Run Time 5.2 Performance Comparison 6 Conclusion A Appendix - KSP Algorithm References Drone Base Stations Transmission Power Control and Localization 1 Introduction 2 Related Work 3 System Model 3.1 Channel Model 3.2 Problem Formulation 4 Expectation-Maximization Algorithm 4.1 Standard EM 4.2 EM-Based Clustering Algorithm 4.3 On Outliers and the Choice of Sth 4.4 On Collision Between DBSs 5 Monte Carlo Tree Search 5.1 Transmission Power MCTS 5.2 Selection Criterion 5.3 Time-Based Updates 6 Simulation and Discussions 6.1 Scenario 1 (PSO) 6.2 Scenario 2 (RWP) 7 Conclusions References Joint Edge Resource Allocation and Path Planning for Drones with Energy Constraints 1 Introduction 2 Related Work 3 System Model and Problem Formulation 3.1 Drones, Paths and Flight Model 3.2 Sensing and Computation 3.3 Energy Model 3.4 Problem 4 Algorithm 4.1 Overview 4.2 Description 4.3 Complexity 5 Evaluation 5.1 Topology and Missions 5.2 Drone and Server Model Parameters 5.3 Settings for EPPOS Parameters 5.4 Presentation of Results 5.5 Experiments for Different Battery Switching Delays 5.6 Experiments for Different Degrees of Autonomy 6 Conclusion References Wireless Networks Range and Capacity of LoRa 2.4 GHz 1 Introduction 2 Background on LoRa 2.4GHz 3 Related Work 3.1 Performance of the Standard LoRa 3.2 Performance of LoRa 2.4GHz 4 Simulator of LoRa 2.4GHz 4.1 Channel Propagation Model 4.2 Fading Models 4.3 Collision Model 4.4 Simulation Setup 5 Simulation Results 5.1 Validation: Comparison with Reported Measurements 5.2 Case of SNR Boundaries and Maximal Load in the Standard LoRa 5.3 Analysis of LoRa 2.4GHz Capacity 6 Conclusion References Reception of Terrestrial DAB+ and FM Radio with a Mobile Device: A Subjective Quality Evaluation 1 Introduction 2 Digital Radio Market 3 About the Study 3.1 Multiplex Configuration 3.2 Tested Devices 3.3 Testing Procedure 4 Results 5 Conclusions References Federated Learning-Based Interference Modeling for Vehicular Dynamic Spectrum Access 1 Introduction 2 Stand-Alone System Model 2.1 Capture of Samples 2.2 Channel Quality Assessment 2.3 Interference Modeling 3 REM and Federated Learning for Interference Modeling 3.1 REM for Interference Modeling 3.2 Local REM Update 3.3 Global REM Update 3.4 Federated Learning Cycle 4 Simulation Scenario 5 Results 5.1 Baseline Model 5.2 Single Platoon Scenario 5.3 Multiple Platoons Scenario 5.4 Channel Selection 6 Conclusion References A Study of Extensive LoRaWAN Downlink Communication in a Mobility Scenario 1 Introduction 2 Background Information 2.1 An Overview of LoRaWAN 2.2 Network Simulator 3 2.3 Related Work 3 Extensive Downlink Communication in LoRaWAN 4 Real World Data Collection 4.1 Test Scenario 1 - Out of Signal 4.2 Test Scenario 2 - Periodic Acknowledgement 5 Simulation of LoRaWAN Class C 6 Conclusion and Future Work References Mobile and Human Computer Interactions Microbenchmarking Documents in the Air 1 Introduction 2 System Architecture 2.1 Usage Scenarios 3 Dissimilarity of Signatures 3.1 Dissimilarity Measure 4 Dissimilarity Microbenchmarks 4.1 Building 1 Results 4.2 Building 2 Results 4.3 Building 3 Results 5 App and Server Implementation 6 Discussion 6.1 Future Work 7 Related Work 8 Conclusion References InnerEye: A Tale on Images Filtered Using Instagram Filters - How Do We Interact with them and How Can We Automatically Identify the Extent of Filtering? 1 Introduction 2 Related Work 2.1 Social and News Media, and Edited Images 2.2 Image Forgery and Fake Image Detection 3 Methodology of Our Study 4 Perception About Interacting with Alterations of Shared Filtered Images 4.1 Surveying Users About Their Experiences with Interacting Through Images on Social Media 4.2 Surveying Social Media Users About Their Preferences Towards Interacting with Types of Image Filtering 4.3 Surveying Capability of General Users in Quantifying Extent of Image Alteration 4.4 Consolidated Survey to Verify Our Findings Across Demographic and Temporal Variations 5 Implementation of an Automated Solution to Detect Extent of Image Alteration: InnerEye 5.1 Developing a Neural Network Model for Detecting Filtered Images 5.2 Developing a Web Tool to Deploy the Neural Network Model 6 Evaluation of Performance of Our Solution 7 Discussion 7.1 Users and Filtered Images 7.2 Variability in General Perception Towards Filtered Images 7.3 Appreciation of Automated Tools in Detecting Filtered Images 7.4 Limitations of Our Study 8 Avenues for Further Research 9 Conclusion References An Interactive Visualization System for Streaming Data Online Exploration 1 Introduction 2 Related Work 3 Challenges and Motivation 3.1 Limitations of Data Cube 3.2 Motivation 4 Design of Linkube 4.1 Basic Definitions 4.2 Linked List-Optimized Structure 4.3 Smart Caching Mechanism 5 Algorithm 5.1 Building the Linked List 5.2 Aggregation State Updating 6 Evaluation 6.1 Implementation 6.2 Experimental Setup 6.3 Effectiveness of Linkube 6.4 Performance on Streaming Data 7 Conclusion References RF-Eye: Training-Free Object Shape Detection Using Directional RF Antenna 1 Introduction 2 Background 3 Methodology 3.1 Challenge 3.2 Our Design Principles 3.3 Design Overview 3.4 Linear Frequency Modulated (LFM) Signal Implementation 3.5 Reduce the Effect of Sidelobe 3.6 Modulation and Demodulation 3.7 Algorithm to Capture Target Shape 3.8 Operator Selection to Detect Image Edge 4 Evaluation 4.1 Experiment Setup and Radio Implementation 4.2 Result of Capturing Target Shape 4.3 The Impact of Sampled Signal Pulse Number 4.4 Impact of Window Functions 4.5 Edge Detection Algorithm Comparison 4.6 Latency 5 Discussion 5.1 How to Choose Carrier Frequency 5.2 Antenna Rotation 6 Conclusion and Future Work References Technology for Health Identification of Abnormal Behavior in Activities of Daily Life Using Novelty Detection 1 Introduction 2 Related Work 3 Methodology 3.1 Pre-processing 3.2 Machine Learning for Novelty Detection 4 Results 5 Conclusion References Behavioral Anomaly Detection of Older People Living Independently 1 Introduction 2 Related Work 3 Data Gathering and Preprocessing 3.1 Data Gathering 3.2 Data Integration 3.3 Incomplete Data Removal 3.4 Data Aggregation 3.5 Data Formatting 4 Analysis of Results 4.1 Analysis of Events Per Hour 4.2 Occupation Probability 4.3 Outlier Detection 5 Discussion of Results 6 Conclusion References Making Everyday Objects More Inclusive: A Case Study via Remote Participatory Design 1 Introduction 2 Related Work 3 Method 4 Participatory Design: Case of the Barbanera Calendar 4.1 Barbanera Calendar 4.2 Designing the First Prototype 4.3 The Pilot Test and Second Design Iteration 4.4 Final Prototype 5 Discussion 6 Design Suggestions 7 Conclusions References The Journey Through Illness of the Type 1 Diabetes Caregiver, from Disruption to Positive Coping 1 Introduction 2 What T1D is 3 Related Work and Research Gap 4 Method 5 The Hero’s Journey in Mythology 6 The Journey 6.1 The Departure 6.2 The Road of Trials 6.3 The Victory 6.4 The Acquisition of New Powers 7 Discussion 8 Conclusion References Author Index