[Internet of Things] Convergence of Artificial Intelligence and the Internet of Things ||

This book gathers recent research work on emerging Artificial Intelligence (AI) methods for processing and storing data generated by cloud-based Internet of Things (IoT) infrastructures. Major topics covered include the analysis and development of AI-powered mechanisms in future IoT applications and architectures. Further, the book addresses new technological developments, current research trends, and industry needs. Presenting case studies, experience and evaluation reports, and best practices in utilizing AI applications in IoT networks, it strikes a good balance between theoretical and practical issues. It also provides technical/scientific information on various aspects of AI technologies, ranging from basic concepts to research grade material, including future directions. The book is intended for researchers, practitioners, engineers and scientists involved in the design and development of protocols and AI applications for IoT-related devices. As the book covers a wide range of mobile applications and scenarios where IoT technologies can be applied, it also offers an essential introduction to the field. Introduction Research Solutions Conclusion References Contents Fog Computing: Data Analytics for Time-Sensitive Applications 1 Introduction 2 Fog Computing Applications 3 Architecture of Fog Computing 3.1 Smart Layer 3.2 Fog Layer 3.3 Cloud Services 4 Benefits of Fog Computing 5 Challenges of Fog Computing 6 Conclusion and Discussions References Medical Image Watermarking in Four Levels Decomposition of DWT Using Multiple Wavelets in IoT Emergence 1 Introduction 2 Digital Image Watermarking Algorithms 2.1 Biorthogonal Wavelet 2.2 Reverse Biorthogonal Wavelet 2.3 Symlet Wavelet 2.4 Coiflets Wavelet 2.5 Discrete Meyer Wavelet 3 The Proposed Medical Image Watermarking Algorithm 4 Experimental Results and Evaluation 5 Conclusion References Optimised Statistical Model Updates in Distributed Intelligence Environments 1 Introduction 1.1 Problem Statement 1.2 Paper Organisation 2 Related Work and Background 2.1 Optimised Sequential Decision Making 2.2 Contribution 3 Methodology 3.1 Optimal Postponing Policy 3.2 Policies Under Comparison 4 Performance Evaluation 4.1 Data Sets 4.2 Experimentation with Linear Regression Models 4.3 Experimentation with Support Vector Regression Models 4.4 Evaluation Summary 5 Conclusions References Intelligent Vehicular Networking Protocols 1 Introduction 2 Routing Protocols in VANET 2.1 Topology Based Routing Protocols 2.2 Position-Based Routing Protocols or Geographic Routing Protocols 2.3 Broadcast Routing 2.4 Geocast Routing Protocols 2.5 Cluster-Based Routing Protocols 3 Internet of Vehicles 3.1 Unicast Protocol 3.2 Multicast Protocol 3.3 Broadcast Protocol 4 Conclusion References Towards Ubiquitous Privacy Decision Support: Machine Prediction of Privacy Decisions in IoT 1 Introduction 2 Related Work 2.1 Prediction of Privacy Decision-Making 2.2 Privacy Segmentation 3 Dataset of Privacy Decisions 4 Privacy Decision Prediction 4.1 Machine Learning Model 4.2 Features 4.3 Training Strategy 4.4 Implications 5 Discussion and Future Work 5.1 Representability of Data 5.2 Reliability of Privacy Segmentation 5.3 Privacy Paradox 6 Conclusion References Energy-Efficient Design of Data Center Spaces in the Era of IoT Exploiting the Concept of Digital Twins 1 Introduction 2 Related Work 2.1 Energy Consumption in Data Centers 2.2 Degrees of Freedom in Energy Efficiency 2.3 Power Usage Effectiveness 2.4 Data Centers and Building Envelopes 3 Proposed Methodology 3.1 Modeling Procedure 3.2 Software Simulation Tools 3.3 Creating the 3D Geometry in SketchUp 3.4 Measurements 3.5 Setting Model Parameters in OpenStudio .OSM File 3.6 Hardware Laboratory’s Simulation Results 3.7 Data Center Simulation Results 4 Models’ Validation 4.1 Hardware Laboratory’s Validation Results 4.2 Data Center’s Validation Results 4.3 PUE Calculations for Various Structural Interventions 5 Conclusions References In-Network Machine Learning Predictive Analytics: A Swarm Intelligence Approach 1 Introduction 1.1 Motivation 1.2 Aim 1.3 Outline 2 Background 2.1 Particle Swarm Optimisation Algorithm 2.2 Particle Representation 2.3 Parameters 2.4 The PSO Algorithm 2.5 Regression 2.6 Prediction Error Metrics 2.7 Network Modelling 2.8 Mica2 Wireless Sensor Platform 3 Analysis 3.1 Problem Definition and Analysis 3.2 Baseline Methodology 3.3 Limitations 3.4 Proposed Methodology 3.5 Network Model Impact 4 Implementation 4.1 Programming Language 4.2 Data 4.3 Particle Swarm Optimisation 4.4 Convergence 4.5 Network Models 5 Performance and Comparative Assessment 5.1 Prerequisites 5.2 Random Network Assessment 5.3 Small World Network Assessment 6 Conclusions and Future Work 6.1 Assessment Results 6.2 Future Work 6.3 Summary References Machine Learning Techniques for Wireless-Powered Ambient Backscatter Communications: Enabling Intelligent IoT Networks in 6G Era 1 Introduction to Artificial Intelligence (AI) 2 Machine Learning Paradigm and Techniques 2.1 Supervised Learning 2.2 Unsupervised Learning 2.3 Reinforcement Learning 3 Robustness of Deep (Machine) Learning Approaches 4 Hardware Requirements for Implementing Machine Learning Techniques 4.1 Computational Cost: CPU Vs GPU 4.2 Existing Hardware Solutions 5 Ambient Backscatter Communications and Machine Learning 5.1 Basics of Backscatter Communications 5.2 Applications of Machine Learning in Ambient Backscatter Communications 5.3 Selection of Reflection Coefficient 6 System Model 7 Power Control in Ambient Backscatter Communications 8 Simulation Results and Discussion 9 Conclusion and Future Research Directions References Processing Systems for Deep Learning Inference on Edge Devices 1 Introduction 2 Deep Learning 3 Complexity Reduction of Deep Learning Models 4 Deep Learning Computation 5 Computing Platforms for Edge Devices 5.1 Reconfigurable Platforms for Edge Computing 5.2 ASICs for Edge Computing 6 Inference at the Edge: Present and Future 7 Conclusions References Power Domain Based Multiple Access for IoT Deployment: Two-Way Transmission Mode and Performance Analysis 1 Introduction 2 System Model 3 Performance Evaluation: Outage Performance Analysis 3.1 The Outage Probability of the First User Pair 4 Numerical Results 5 Conclusion References Big Data Thinning: Knowledge Discovery from Relevant Data 1 Introduction 1.1 Motivation 1.2 Aims and Hypotheses 1.3 Performance Assessment 1.4 Outline 2 Background and Relevant Research 2.1 Competitive Learning 2.2 Learning Automata with Two Actions 2.3 The Improved Initialisation Algorithm: K-Means++ 3 Hypothesis 1 3.1 Design 3.2 Implementation 3.3 Evaluation 4 Hypothesis 2 4.1 Design 5 Implementation 5.1 Generating Queries/Building the Query Space 5.2 Query Quantisation Using RPCL 5.3 Evaluation 6 Conclusions and Future Work 6.1 Generalisation of Findings 6.2 Future Work References Optimizing Blockchain Networks with Artificial Intelligence: Towards Efficient and Reliable IoT Applications 1 Blockchain 1.1 Blocks 1.2 Transaction 1.3 Signing a Transaction 1.4 Mining 2 Applications of Blockchain Technology 2.1 Local Energy Trading 2.2 Internet of Things 2.3 Next Generation Payment Solutions 3 Fundamentals of Artificial Intelligence 3.1 Artificial Intelligence in Modern Era 3.2 Impact of Artificial Intelligence on Industries 4 Data and Infrastructure Desideratum of Artificial Intelligence 4.1 IT Infrastructures 4.2 Algorithms and Methods 4.3 Training Data 5 Blockchain Network Architecture 6 System Model 7 Optimization Using Deep Neural Networks 7.1 Problem Formulation 7.2 Deep Learning Network Setup 8 Numerical Results 9 Conclusion References Industrial and Artificial Internet of Things with Augmented Reality 1 Introduction 1.1 Artificial Intelligence 1.2 Augmented Reality 2 Industrial Internet of Things 2.1 Cyber-Physical Systems 2.2 Industrial Augmented Reality 2.3 IoT Communication Systems 3 Artificial Internet of Things 3.1 Bringing Intelligence to the Data 3.2 Edge and Fog Computing 4 Applications 4.1 Manufacturing Sector 4.2 Aerospace Sector 4.3 Logistics Sector 4.4 Water Sector 5 Future Directions 6 Conclusion References IoT Detection Techniques for Modeling Post-Fire Landscape Alteration Using Multitemporal Spectral Indices 1 Introduction 1.1 Remote Sensing: Delivering on the IoT 1.2 Study Area 2 Materials and Methods 2.1 Area of Interest 2.2 Remote Sensors on Landsat 8 OLI/TIRS 2.3 Dataset 2.4 Classification 2.5 Variation Detection Processing 3 Results Verification 3.1 Results 3.2 Significance of Results 4 Conclusion References Internet of Things and Artificial Intelligence—A Wining Partnership? 1 Introduction 2 Technology Background—Five Decades that Changed the Electrical Engineering Paradigm 3 Human Background 4 Challenges of IoT-AI Partnership 5 Towards an Intelligent Strategy for AI-Driven IoT. Areas of Urgent Research 5.1 Engineering Education 5.2 Design for Accountability (DfA) 6 Conclusions References AI Architectures for Very Smart Sensors 1 Introduction 2 Key Characteristics of Artificial Neural Networks 2.1 Hierarchy of Feature Detectors 2.2 Fully Connected Layers in Neural Networks 2.3 Convolutional Layers in Neural Networks 2.4 Vanishing or Exploding Gradient in Connection with Stochastic Gradient Descent 2.5 Shortcut Connections 2.6 Transfer Learning 3 Architectures of Modern Neural Networks 3.1 LeNet-5 3.2 AlexNet 3.3 VGGNet 3.4 GoogLeNet 3.5 ResNet 3.6 U-Net 3.7 DenseNet 3.8 PolyNet 3.9 ResNeXt 3.10 DualPathNet 4 Classification Accuracy of Modern Neural Networks 5 Computational Complexity Reducing Techniques of Modern Neural Networks 5.1 Reducing the Depth and Width of Neural Networks 5.2 Reducing the Resolution of the Input 5.3 More Efficient Building Blocks of Neural Networks 5.4 Separable Convolution 5.5 Depth-Wise Separable Convolution 5.6 Pruning 5.7 Compression of Parameters 5.8 Reduced Floating-Point Precision 5.9 Quantification of Weights into Integer or Binary Values 5.10 Quantification of Both Weights and Activations into Integer or Binary Values 5.11 More Efficient Computational Hardware 5.12 Custom Hardware Design 6 Neural Networks for Mobile Devices and IoT 6.1 SqueezeNet 6.2 MobileNet 6.3 ShuffleNet 6.4 LQ-Nets 6.5 GroupNet 7 Classification Accuracy of Neural Networks for Mobile and IoT 8 Conclusion References