Mobile Computing, Applications, and Services: 12th EAI International Conference, MobiCASE 2021, Virtual Event, November 13–14, 2021, Proceedings ... and Telecommunications Engineering)

This book constitutes the thoroughly refereed post-conference proceedings of the 12th International Conference on Mobile Computing, Applications, and Services, MobiCASE 2021, held in November 2021. Due to COVID-19 pandemic the conference was held virtually. The 9 full papers were carefully reviewed and selected from 21 submissions. The papers are organized in two topical tracks: mobile application and deep learning, and mobile application with data analysis. Preface Organization Contents Mobile Application and Deep Learning YOLO-RFB: An Improved Traffic Sign Detection Model 1 Introduction 2 YOLO V4 3 Improved YOLO V4 3.1 RFB 3.2 YOLO-RFB 4 Experiment and Analysis 4.1 Common Data Sets 4.2 Experimental Environment and Hyperparameter Setting 4.3 Evaluation Indicators 4.4 Experimental Results and Comparative Presentation 5 Conclusion References Privacy-Preserving Sharing of Mobile Sensor Data 1 Introduction 2 Go-Between Overview 2.1 Typical Application Scenarios 2.2 Solution Overview 2.3 Threat Model 2.4 Enabling Theory and Technologies 3 Applying and Complementing DP 3.1 From Theory to Practice 3.2 Privacy and Utility Tradeoffs 3.3 Data Contribution Incentives 4 System Design and Implementation 4.1 Architecture 4.2 Privacy Mechanism 5 Evaluation 5.1 Experimental Setup 5.2 Evaluation Design 5.3 Results 6 Discussion 7 Related Work 8 Conclusion References Service Performance Analysis of Cloud Computing Server by Queuing System 1 Introduction 2 Cloud Computing Server Queuing System Model 3 Model Analysis and Service Performance Metrics by Geo/G/1 4 Service Performance Metrics by Geo/G/1with the Vacation 5 Model Analysis and Service Performance Index 5.1 Service Performance Index with no Vacation 5.2 Service Performance Index with Vacation 5.3 Optimizing Concurrent Number of the Cloud Computing References CLES: A Universal Wrench for Embedded Systems Communication and Coordination 1 Introduction 2 Problem Domain: Embedded Distributed Computing 2.1 Use Case: Vehicle Platooning 2.2 Use Case: Joint All-Domain Command and Control (JDAC) 3 RESTful Architecture for Distributed Embedded Systems 3.1 Requirements 3.2 CLES Core Language Design 3.3 CLES SDK Interface Definition 3.4 CLES Point-to-Point Service 3.5 CLES Runtime and Task Distribution 3.6 CLES Implementation 4 Evaluation 4.1 P2P CLES Performance and Micro-Benchmarks 4.2 CLES Runtime Evaluation 5 Conclusion References When Neural Networks Using Different Sensors Create Similar Features 1 Introduction 2 Problem Position 2.1 Deep Feature Extraction 2.2 Canonical Correlation Analysis 2.3 A Simplistic Example 2.4 Extensions 3 Related Works 3.1 CCA as a Fusion Method 3.2 Similarity of Different Neural Networks 3.3 SVCCA and Improvements 4 Experiments 4.1 Datasets 4.2 Studying Component Similarity with Subspace Projection 5 Conclusion References Mobile Application with Data Analysis Improving Togetherness Using Structural Entropy 1 Introduction 1.1 Background 1.2 Contributions 1.3 Related Work 2 Togetherness Based on Structural Entropy 2.1 Network Integration and Togetherness 2.2 Structural Entropy 2.3 Togetherness Improvement Through Entropy 3 The TIE Algorithm 4 Experiment 4.1 Experimental Settings 4.2 Results and Analysis 5 Conclusion and Future Work References Blockchain-Based Group Key Agreement 1 Introduction 2 Related Works 3 Preliminaries 3.1 Data Format of Bitcoin Transactions 3.2 Group Key Agreement 4 Protocols 4.1 A Blockchain-Based Two-Party Key Agreement Protocol 4.2 A Blockchain-Based Group Key Agreement Protocol 5 Security Analysis 5.1 The Informal Security Analysis for the Session Key 5.2 The Informal Security Analysis for the Blockchain Addresses 5.3 The Informal Security Analysis for Forward Secrecy and Backward Secrecy 5.4 Reply Attacks 5.5 Man-in-the-Middle Attacks 6 Conclusion and Future Work References Attacking Community Detectors: Mislead Detectors via Manipulating the Graph Structure 1 Introduction 2 Related Work 3 Problem Formulation 3.1 Attacker's Model 3.2 A Form of Topology Attack 4 Targeted Community Attack (TCA) Based on Entropy 5 Experiments 5.1 Data 5.2 Benchmark, Community Detectors and Metrics 5.3 Experimental Settings and Results 6 Conclusion References ResNet-Like CNN Architecture and Saliency Map for Human Activity Recognition 1 Introduction 2 Related Work 2.1 CNNs for HAR 2.2 ResNet 2.3 Saliency Map 3 ResNet-Like Structure 3.1 Residual Learning 3.2 Network Architecture 4 Saliency Map 5 Experiment and Result 5.1 Dataset 5.2 Experiment Parameters 5.3 Performance 5.4 Discussion of Unbalance 5.5 Saliency Map Usage 6 Conclusion and Future Work References Author Index