Predictive Analytics in Cloud, Fog, and Edge Computing: Perspectives and Practices of Blockchain, IoT, and 5G

Preface
Acknowledgement
Contents
Collaboration of IoT and Cloud Computing Towards HealthcareSecurity
	1 Introduction
	2 Inspiration
	3 Related Work and Background
	4 Cloud Computing Deployment Models
		4.1 Public Internet
		4.2 Corporate Cloud
		4.3 Cloud Hybrid
		4.4 Cloud Provider
	5 Utility Computing Service Models
		5.1 Software as a Service (SaaS)
		5.2 Infrastructure as a Service (IaaS)
		5.3 Platform as a Service (PaaS)
	6 Security Issues
	7 Threats in Cloud Computing
		7.1 Compromised Identities and Broken Security
		7.2 Data Infringement
		7.3 Hacked Frontier and APIs
		7.4 Manipulated System Vulnerabilities
		7.5 Permanent Data Loss
		7.6 Inadequate Assiduity
		7.7 Cloud Service Inattention
		7.8 DoS Attacks
		7.9 Security Challenges in Cloud Infrastructure
			7.9.1 Security Challenges
			7.9.2 Challenges of Deployed Models
			7.9.3 Resource Pooling
			7.9.4 Unencrypted Data
			7.9.5 Identity Management and Authentication
			7.9.6 Network Issues
		7.10 Point at Issue in the IoT Health Care Framework
			7.10.1 Reliability
			7.10.2 Discretion
			7.10.3 Solitude
			7.10.4 Unintended Efforts
		7.11 Challenges
			7.11.1 Security
			7.11.2 Confidentiality
			7.11.3 Assimilation
			7.11.4 Business Illustration
		7.12 Dispensing Refined Patient Supervision
		7.13 Character of IoT in Healthcare
		7.14 Conclusion
		7.15 Future Work
	References
Robust, Reversible Medical Image Watermarking for Transmission of Medical Images over Cloud in Smart IoT Healthcare
	1 Introduction
	2 Related Work
	3 Proposed Work
		3.1 EHR Insertion (Embedding) and Retrieval (Extraction)
		3.2 EHR Encryption and Decryption
	4 Experimental Results and Discussion
	5 Conclusions
	References
The Role of Blockchain in Cloud Computing
	1 Blockchain
		1.1 Introduction
		1.2 Characteristics
			1.2.1 Immutability
			1.2.2 Distributed
			1.2.3 Enhanced Security
			1.2.4 Distributed Ledgers
			1.2.5 Faster Settlement
			1.2.6 Working of Blockchain
		1.3 Major Implementations
			1.3.1 Cryptocurrencies
			1.3.2 Smart Contracts
			1.3.3 Monetary Services
			1.3.4 Games
		1.4 Blockchain Types
		1.5 There Are Mainly 4 Types of Blockchain as Shown in Table 1
			1.5.1 Public Blockchain Networks
			1.5.2 Exclusive Blockchain Networks
			1.5.3 Hybrid Blockchain Networks
			1.5.4 Consortium Networks
		1.6 Advantages
			1.6.1 Secure
			1.6.2 There Will Be No Intervention from Third Parties
			1.6.3 Safe Transactions
			1.6.4 Automation
		1.7 Disadvantages
			1.7.1 High Implementation Cost
			1.7.2 Incompetency
			1.7.3 Private Keys
			1.7.4 Storage Capacity
	2 Cloud Computing
		2.1 What Is Cloud Computing?
		2.2 Deployment Models in Cloud
			2.2.1 Public Cloud
			2.2.2 Private Cloud
			2.2.3 Hybrid Cloud
			2.2.4 Community Cloud
		2.3 Implementations of Cloud Computing
			2.3.1 Web Based Services
			2.3.2 Software as a Service
			2.3.3 Infrastructure as a Service
			2.3.4 Platform as a Service
		2.4 Comparison of Cloud Computing Model with Traditional Model
			2.4.1 Persistency
			2.4.2 Automation
			2.4.3 Cost
			2.4.4 Security
		2.5 Advantages of Cloud Computing
			2.5.1 Cost Efficiency
			2.5.2 Backup and Recovery
			2.5.3 Integration of Software
			2.5.4 Information Availability
			2.5.5 Deployment
			2.5.6 Easier Scale for Services and Delivery of New Services
		2.6 Challenges of Cloud Computing
			2.6.1 Technical Problems
			2.6.2 Certainty
			2.6.3 Vulnerable Attacks
			2.6.4 Suspension
			2.6.5 Inflexibility
			2.6.6 Lack of Assistance
		2.7 Integration of Cloud Computing with Block Chain
			2.7.1 The Advantages of Combining Cloud and Blockchain Technology
			2.7.2 Blockchain Support for Cloud Computing
			2.7.3 Deduplication of Data in the Cloud with Blockchain
			2.7.4 Access Control Based on Blockchain in Cloud
	References
Analysis and Prediction of Plant Growth in a Cloud-Based Smart Sensor Controlled Environment
	1 Introduction
	2 Literature Survey
	3 IoT in Greenhouse
		3.1 Architecture
		3.2 Cloud Implementation
		3.3 Hardware Components (Fig. 2)
	4 System Overview
		4.1 Dataset
		4.2 Data Preprocessing
		4.3 LightGBM
		4.4 Training and Building the Model
	5 Results and Explanation
	6 Conclusion
	References
Cloud-Based IoT Controlled System Model for Plant DiseaseMonitoring
	1 Introduction
	2 Literature Survey
	3 IoT Controlled Device
	4 Cloud Architecture
	5 Methodology
		5.1 HOG Filter
	6 Experimental Analysis
		6.1 Analysis Using Artificial Neural Network
		6.2 Analysis Using Convolutional Neural Network
	7 Conclusion
	References
Design and Usage of a Digital E-Pharmacy Application Framework
	1 Introduction
	2 Literature Survey
	3 Utilization of Cloud in Health Care
	4 Redefining E-Pharmacy Domain
	5 Impact of Cloud Computing in Pharmacy
	6 Model Design and Implementation
	7 Basic Structure of the Cloud Based E-Pharmacy Application
	8 Security Provided by the Application
		8.1 XSS Security (Cross Site Scripting)
		8.2 CSRF Token (Cross Site Request Forgery)
		8.3 SQL Injection Security
		8.4 User Upload Security
	9 Results and Discussion
	10 Important Features of the Application
	11 Critical Goals of the Application
	12 Benefits of the Model
	13 Summary/Conclusion
	References
Serverless Data Pipelines for IoT Data Analytics: A Cloud Vendors Perspective and Solutions
	1 Introduction
		1.1 Motivation
		1.2 Contributions
	2 Background
		2.1 Internet of Things
		2.2 Serverless Data Pipelines for IoT Data Processing
	3 Literature Survey
	4 Cloud Service Providers (CSP) and IoT Solutions
		4.1 Edge Tier
			4.1.1 Comparison of AWS IoT Greengrass and Azure IoT Edge
		4.2 Cloud Tier
	5  Real-Time IoT Application: Predictive Maintenance of Industrial Motor
	6 Building SDP for Predictive Maintenance Application
		6.1 Proposed Serverless Data Pipelines
			6.1.1 Building an Anomaly Detection Model
		6.2 SDP Using AWS and Microsoft Azure
	7 Experiments and Results
		7.1 Performance Metrics
		7.2 Experimental Setup
		7.3 Results and Discussions
	8 Conclusions
	References
Integration of Predictive Analytics and Cloud Computing for Mental Health Prediction
	1 Introduction
	2 Method of Approach
		2.1 Overview of the Subject
			2.1.1 Supervised Learning
			2.1.2 Unsupervised Learning
		2.2 Selection of Papers
		2.3 Literature Search Strategy
		2.4 Study Selection
		2.5 Data Extraction and Analysis
	3 Introduction to Mental Health Research
		3.1 Machine Learning in Big Data
		3.2 Deep Learning in Healthcare
		3.3 Natural Language Processing
	4 The Pipeline of Data Flows from the Sensors to the Algorithmic Approach
		4.1 Sensor Data
		4.2 Extraction of Features
		4.3 Designing the Behavioural Markers
		4.4 Clinical Target
	5 Cloud Computing
		5.1 Architecture of Cloud Computing
		5.2 Benefits of Cloud Computing in the Healthcare Industry
		5.3 Cloud Computing as a Solution to Mental Health Issues
	6 Review of Personal Sensing Research
	7 Result of the Research
		7.1 Limitations of the Study Done on the Algorithms to Detect Mental Health
		7.2 Results Based on iCBT Test
	8 Discussion
	9 Conclusion
	References
Impact of 5G Technologies on Cloud Analytics
	1 Introduction
	2 Self-Organizing Next Generation Network Data Analytics in the Cloud
		2.1 What Is Network Data Analytics?
		2.2 Benefits of Network Data Analytics
		2.3 The Best Uses of Network Data Analytics
		2.4 The Near Future
		2.5 The Opportunities
	3 Intelligent 5G Network Estimation Techniques in the Cloud
		3.1 Network Estimation Technique
		3.2 Literature Review
	4 5G-cloud Integration: Intelligent Security Protocol and Analytics
		4.1 Scope
		4.2 5G Cloud Threat
		4.3 5G-Cloud Integration
		4.4 Advantages of Security Capabilities
	5 5G, Fog and Edge Based Approaches for Predictive Analytics
		5.1 Introduction
		5.2 Literature Review
	6 5G and Beyond in Cloud, Edge, and Fog Computing
		6.1 Edge Computing
		6.2 Cloud Computing
		6.3 5G and Beyond
	7 AI-Enabled Next Generation 6G Wireless Communication
		7.1 Computation Efficiency and Accuracy
		7.2 Hardware Development
		7.3 Types 6 G Wireless Communication
		7.4 6G Wireless Access Use Case
	References
IoT Based ECG-SCG Big Data Analysis Framework for Continuous Cardiac Health Monitoring in Cloud Data Centers
	1 Introduction
	2 Related Work
	3 Proposed Cardiac Big Data Analysis Framework
		3.1 ECG/SCG Data Collection Framework
		3.2 Data Processing and Analysis Framework
		3.3 MapReduce Based Cardiac Big Data Processing Model
	4 Evaluation Results
	5 Conclusion and Future Works
	References
A Workload-Aware Data Placement Scheme for Hadoop-Enabled MapReduce Cloud Data Centers
	1 Introduction
	2 Related Works
	3 Problem Description
	4 Proposed Protocol
		4.1 System Model
	5 Problem Formulation
		5.1 Network Model
		5.2 Task Processing Model
		5.3 Workload Distribution
	6 Data Locality Problem
	7 Conclusion and Future Works
	References
5G Enabled Smart City Using Cloud Environment
	1 Introduction
	2 Technologies Used to Build the Smart City
		2.1 Edge and Fog Computing
		2.2 What Price Does 5G Provide for Fog Computing?
		2.3 Cloud Computing
		2.4 Internet of Things
	3 SmartCity Architecture
	4 Smart City Service Cases
		4.1 Smart Grid
		4.2 Smart Healthcare
		4.3 Smart Transport
		4.4 Smart Governance
		4.5 Remote Monitoring
		4.6 Event Detection
		4.7 Emergency Response
		4.8 Emotional Monitoring
		4.9 Crowd Management
		4.10 Flexible Building Materials
		4.11 Environmental Monitoring
		4.12 Smart Electrical Power Distribution
		4.13 Smart Precision Agriculture
		4.14 Animal Health Monitoring System
	5 Case Study of Smart City
		5.1 Barcelona
		5.2 Smart Dubai Happiness Meter – Dubai, United Arab Emirates (UAE)
		5.3 #SmartME
		5.4 Urban Area Quality Index – Russian Federation
	6 Challenges and Problems
		6.1 Business Challenges
			6.1.1 Planning
			6.1.2 Stability
			6.1.3 Market Source and Customer
			6.1.4 Smart City Acquisition Costs
			6.1.5 Cloud Computing Integration
		6.2 Technical Challenges
			6.2.1 Privacy
			6.2.2 Data Analysis
			6.2.3 Data Integration
			6.2.4 Visualization based on GIS
			6.2.5 Quality of Service
			6.2.6 Computational Intelligence Algorithms for Smart City Big Data Analytics
	7 Conclusion
	References
Hardware Implementation for Spiking Neural Networkson Edge Devices
	1 Introduction
	2 The Spiking Neural Network (SNN)
		2.1 The Leaky Integrate-and-Fire (LIF) Neuron Model
		2.2 The Learning Algorithms
	3 Hardware Accelerators for SNNs on the Edge
		3.1 Optimizations that Exploit the Temporal Sparsity of SNN
		3.2 Data and Memory-Centric Architectures
		3.3 Flexible Hardware Architectures for SNN on the Edge
	4 Algorithm Design
		4.1 Synapse Pruning
		4.2 Hybrid On/Off Chip Training
		4.3 Quantization and Binarization
		4.4 Time Step Reduction
	5 SNN versus ANN for Edge Computing
		5.1 Memory Consumption
		5.2 Energy Consumption
	6 Conclusions
	References