Smart Sensor Networks: Analytics, Sharing and Control (Studies in Big Data, 92)

✍ Scribed by Umang Singh (editor), Ajith Abraham (editor), Arturas Kaklauskas (editor), Tzung-Pei Hong (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 233
• Category: Library

✦ Synopsis

This book provides IT professionals, educators, researchers, and students a compendium of knowledge on smart sensors and devices, types of sensors, data analysis and monitoring with the help of smart sensors, decision making, impact of machine learning algorithms, and artificial intelligence-related methodologies for data analysis and understanding of smart applications in networks.

Smart sensor networks play an important role in the establishment of network devices which can easily interact with physical world through plethora of variety of sensors for collecting and monitoring the surrounding context and allowing environment information. Apart from military applications, smart sensor networks are used in many civilian applications nowadays and there is a need to manage high volume of demands in related applications.

This book comprises of 9 chapters and presents a valuable insight on the original research and review articles on the latest achievements that contributes to the field of smart sensor networks and their usage in real-life applications like smart city, smart home, e-healthcare, smart social sensing networks, etc. Chapters illustrate technological advances and trends, examine research opportunities, highlight best practices and standards, and discuss applications and adoption. Some chapters also provide holistic and multiple perspectives while examining the impact of smart sensor networks and the role of data analytics, data sharing, and its control along with future prospects.

✦ Table of Contents

Preface
Contents
About the Editors
Smart Sensors and Devices in Artificial Intelligence
1 An Overview of Artificial Intelligence Technology Directed at Smart Sensors and Devices from a Modern Perspective
Abstract
1 Introduction
2 Methodology
3 Artificial Intelligence Concept
3.1 Machine Learning
3.2 Computer Vision
3.3 Deep Learning
3.4 Natural Language Processing (NLP)
3.5 Big Data
4 Smart Sensor and Devices Concepts
4.1 Sensor Applications in Industry 4.0
5 AI Application on the Smart Sensor, Analytics, Sharing, and Control Perspective
5.1 Health Applications
6 Discussion
7 Geospatial AI
8 Trends
9 Conclusions
References
2 The Role of Smart Sensors in Smart City
Abstract
1 Introduction
1.1 Use of Sensor Technology, Remote Control Network, and Smart Mobility
1.2 Smart Sensors
1.3 Need and Use of Smart Sensor Technology
2 Types of Smart Sensors
2.1 Temperature Sensors
2.2 Types of Temperature Sensors
2.2.1 Thermocouples
2.2.2 RTD (Resistance Temperature Detector)
2.2.3 Thermistors
2.2.4 Semiconductor Based Temperature Sensor or Integrated Circuit (IC) Temperature Sensor
2.3 Light Sensors
2.3.1 Photo-Resistors (LDR)
2.3.2 Photodiodes
2.3.3 Phototransistors
2.4 Air Quality Detection Sensors
2.5 Motion/Speed Sensors
2.5.1 Dual Tech/Hybrid
2.5.2 Microwave Motion Sensors
2.5.3 Motion Sensors
2.6 Smart Smog and Carbon Monoxide Detectors
2.7 Smart Plant Sensors
2.8 Smart Climate Sensors
3 The Usage of Smart Sensors: Prominent Examples
3.1 Smart Water Management (SWM) Using Smart Sensors
3.2 Energy Conservation Using Smart Sensors
3.2.1 Benefits of IoT and Smart Sensors for Energy Efficiency
3.3 Smart Sensor Street Lighting Control System
3.4 Smart Waste Management with Smart Sensors
4 Concluding Remarks
5 Future Scope
References
Impact of AI and Machine Learning in Smart Sensor Networks
3 Impact of AI and Machine Learning in Smart Sensor Networks for Health Care
Abstract
1 Introduction
2 Wireless Smart Sensor Networks (WSSN)
3 Brief Introduction to AI Techniques
3.1 Main Categories of AI Techniques
3.2 Neural Network Algorithms in ML and DL
4 Brief Introduction to ML Techniques
4.1 Supervised Learning (SL)
4.2 Unsupervised Learning (UL)
4.3 Semi-supervised Learning (SSL)
4.4 Reinforcement Learning (RL)
5 Scientific Applications of AI and Machine Learning (ML) Techniques
6 Need for Automation in WSSN
7 Impact of ML Techniques in WSSN
8 Healthcare Applications of AI/ML in WSSN
8.1 Medical Image Diagnosis and Analysis
8.2 Prediction of Health Risks
8.3 Health Information Management
8.4 Drug Discovery and Development
8.5 Tumour Detection
8.6 Electronic Medical Record (EMR)
8.7 Immuno-Oncology Research
9 Conclusion
10 Future Scope
References
Machine Learning Algorithms and Methodologies for Smart Sensor Networks
4 ML Algorithms for Smart Sensor Networks
Abstract
1 Introduction
1.1 Basics of Smart Sensor Networks
2 Introduction to ML Approaches
2.1 Supervised Learning
2.2 Unsupervised Learning
2.3 Semi-supervised Learning
2.4 Reinforcement Learning (RL)
3 ML in SSNs
3.1 Operational Challenges
3.2 Non-operational Challenges
4 Future Applications of ML in SSNs
5 Concluding Remarks
References
Data Analysis for Smart Sensor Networks
5 Energy Efficient Smart Lighting System for Rooms
Abstract
1 Introduction
2 Techniques and Tools for Smart Lighting System Sensors
2.1 IoT
2.2 LDR
2.3 Home Intelligence (HI)
2.4 Occupancy-Sensing Based Systems
2.5 Sensor Technology
2.6 Arduino Board
3 Commercial Smart Lighting Systems
4 Overview of Lighting System
5 Review of Literature Methods of Smart Lighting Systems and Various Case Studies
6 Comparative Approach of the Implementation Mechanisms
6.1 Smart Home System Based on DTMF Technology
6.2 Smart Home System Based on GSM
6.3 Smart Home System Based on Voice Recognition
6.4 Smart Home System Based on Wi-Fi and Internet
6.5 Configuration and Implementation of Smart Lighting System
7 Traditional Lighting System
8 Smart Lighting System
9 Proposal of a Facial Recognition Based Lighting Management System
10 Results for Self-adjusting Lighting System
10.1 Calculations
11 Conclusion
References
6 QUIC Protocol Based Monitoring Probes for Network Devices Monitor and Alerts
Abstract
1 Introduction
2 Background
2.1 TCP and UDP
2.2 TLS and DTLS
2.3 QUIC
3 Traditional Network and QUIC
4 HTTP Based Monitoring Systems
4.1 Host Monitoring Services
5 Need for QUIC in Existing Infra Structure
5.1 Handshake Challenges for Inside Conflict Environment
6 Experimental Setup and Evaluation
7 Conclusions
References
7 External Threat Detection in Smart Sensor Networks Using Machine Learning Approach
Abstract
1 Introduction
2 Background and Related Works
3 Proposed Model
3.1 Algorithm and Process Flow of the Proposed Intrusion Detection Method
3.2 Process Flow of the Proposed Methodology
4 Methodology
4.1 Data Collection and Dataset Analogy
4.2 Data Transformation
4.3 Classification of Attacks
4.4 Data Analysis
4.5 Feature Engineering and Building the Training and the Testing Sets
4.6 Building Out the Predictive Models
5 Comparative Analysis and Model Evaluation
5.1 Comparative Analysis of Data
5.1.1 Analysis of Attacks on the Basis of Protocol as a Parameter
5.1.2 Analysis of Attacks on the Basis of Flag as a Parameter
5.1.3 Analysis of Attacks on the Basis of Service as a Parameter
5.2 Model Evaluation
6 Performance Analysis
7 Future Research Directions
8 Conclusion
References
8 Towards Smart Farming Through Machine Learning-Based Automatic Irrigation Planning
Abstract
1 Introduction
2 Related Works
3 Smart Farming and Irrigation Scheduling
3.1 Smart Farming
3.2 The Water Balance Approach
3.3 The Bowen Ratio-Energy Balance
3.4 The FAO PM Method
3.5 Water Needs Prediction Process
4 The Proposed Framework
4.1 An Overview
4.2 Decision-Making Process
4.3 Materials and Methods
4.3.1 XGBoost
4.3.2 Random Forest
4.3.3 Deep Learning Neural Networks
4.3.4 Evaluation Metrics
4.3.5 Description of Output and Input Variables
4.3.6 Principal Component Analysis
5 Case Study
5.1 Water Requirements for Grain Corn
6 Results and Discussion
6.1 Data Preprocessing: Principal Component Analysis (Study of Variables)
6.1.1 Normalization of Variables (Data Reduction)
6.2 Implementation
7 Conclusion and Perspectives
References
Machine Learning Applications for Smart Sensor Networks
9 Graph Powered Machine Learning in Smart Sensor Networks
Abstract
1 Introduction
2 Literature Survey
3 Framework Based on Graphical Features
3.1 Categorize and Graph Representation of Sensor
3.2 Extraction of Features
3.3 Feature Selection and Classification
3.4 Additional Features
4 Deep Learning
4.1 GCN (Graph Convolutional Network) Approach
4.1.1 Advantages of Graph Convolutional Network
4.1.2 DGCNN (Deep Graph Convolutional Neural Network)
5 Window-Based Approach with Graphical Features
6 Conclusion
7 Future Research
References
Author Index