Computational Intelligence in Recent Communication Networks (EAI/Springer Innovations in Communication and Computing)

✍ Scribed by Mariya Ouaissa (editor), Zakaria Boulouard (editor), Mariyam Ouaissa (editor), Bassma Guermah (editor)

• Publisher: Springer
• Year: 2022
• Tongue: English
• Leaves: 279
• Edition: 1st ed. 2022
• Category: Library

✦ Synopsis

This book focuses on the use of Artificial Intelligence and Machine Learning (AI/ML) based techniques to solve issues related to communication networks, their layers, as well as their applications. The book first offers an introduction to recent trends regarding communication networks. The authors then provide an overview of theoretical concepts of AI/ML, techniques and protocols used in different layers of communication. Furthermore, this book presents solutions that help analyze complex patterns in user data and ultimately improve productivity. Throughout, AI/ML-based solutions are provided, for topics such as signal detection, channel modeling, resource optimization, routing protocol design, transport layer optimization, user/application behavior prediction, software-defined networking, congestion control, communication network optimization, security, and anomaly detection. The book features chapters from a large spectrum of authors including researchers, students, as well as industrials involved in research and development.

✦ Table of Contents

Preface
Contents
About the Editors
1 An Overview of Blockchain and 5G Networks
1.1 Introduction
1.2 Background
1.2.1 Blockchain
1.2.1.1 Blockchain Taxonomy
1.2.1.2 Blockchain Platform Types
1.2.1.3 Blockchain Consensus
1.2.1.4 Blockchain Smart Contract
1.2.1.5 Blockchain Sharding
1.2.1.6 Blockchain Oracle
1.3 5G Networks and Beyond: An Overview
1.3.1 Software-Defined Networking (SDN)
1.3.2 Network Function Virtualization (NFV)
1.3.3 Network Slicing
1.3.4 Multi-Access Edge Computing (MEC)
1.3.5 Device to Device (D2D)
1.3.6 Cloud Computing (CC)
1.4 Blockchain for 5G
1.4.1 Blockchain Integration with 5G Networks
1.4.2 Opportunities Brought by Blockchain Integration with 5G Networks
1.4.2.1 Security Improvements
1.4.2.2 Performance Enhancements
1.5 A Scalable and Secure Blockchain Suitable for 5G
1.5.1 A Scalable and Secure Blockchain Architecture Suitable for 5G
1.5.2 Architecture
1.5.2.1 Shared Blockchain
1.5.2.2 Peer-to-Peer Oracle Network
1.5.3 Design Components
1.5.3.1 Initialization
1.5.3.2 Reward
1.6 Challenges and Future Research Directions
1.6.1 Scalability and Performance
1.6.2 Standardization and Regulations
1.6.3 Resource Constraints
1.6.4 Interoperability
1.6.5 Security
1.6.6 Infrastructure Costs
1.7 Conclusion
References
2 Deep Learning Approach for Interference Mitigation in MIMO-FBMC/OQAM Systems
2.1 Introduction
2.2 MIMO-FBMC/OQAM System Model
2.3 Problem Formulation
2.4 Deep Neural Network for Blind Detection and Interference Mitigation in MIMO-FBMC/OQAM Systems
2.4.1 Data Set
2.4.2 Learning Rule
2.5 Simulation Results
2.5.1 Deep Neural Network Performance: RMSE and Loss
2.5.2 Bit Error Rate
2.6 Conclusion
References
3 Deep Learning-Based Non-orthogonal Multiple Access for 5G and Beyond Networks
3.1 Introduction
3.2 System Model
3.2.1 Principle
3.2.2 NOMA for Downlink
3.2.3 NOMA for Uplink
3.2.4 Imperfection in NOMA
3.2.5 Spectral and Energy Efficiency
3.3 Overview of Deep Learning Models
3.3.1 Deep Neural Network
3.3.2 Convolutional Neural Network
3.3.3 Recurrent Neural Network
3.4 Deep Learning-Based NOMA
3.5 Conclusion
References
4 Traffic Sign Detection: A Comparative Study Between CNN and RNN
4.1 Introduction
4.2 Materials and Methods
4.2.1 Convolutional Neural Networks
4.2.1.1 Multilayer Neural Networks
4.2.1.2 Feed-Forward Neural Network
4.2.1.3 Learning and Training
4.2.1.4 Structure of a Convolutional Neural Network
4.2.1.5 Convolutional Layers
4.2.1.6 Shared Weights
4.2.1.7 Multiple Filters
4.2.1.8 Subsampling Layers
4.2.1.9 Fully Connected Layers
4.2.1.10 Correction Layers (ReLU)
4.2.1.11 Loss Layer (LOSS)
4.2.2 Recurrent Neural Networks
4.2.2.1 Applications of RNNs
4.2.2.2 Loss Function
4.2.2.3 Temporal Back Propagation
4.2.2.4 Activation Functions
4.2.2.5 Long Short-Term Memory
4.3 Proposed System
4.3.1 Comparison Between CNN and RNN
4.3.2 Our System
4.4 Results Obtained
4.5 Conclusion
References
5 Merging Attack-Defense Tree and Game Theory to Analyze Vehicular Ad Hoc Network Security
5.1 Introduction
5.2 System Model
5.3 Fundamental Attack-Defense Tree
5.4 Attack-Defense Tree for VANET Availability
5.5 ROI and ROA for Attack-Defense Tree
5.6 VANET Availability Attack-Defense Game
5.6.1 Basics of the Game Theory
5.6.2 Modeling VANET Availability Attack-Defense Game
5.7 Conclusion
References
6 A Secure Vehicle to Everything (V2X) Communication Model for Intelligent Transportation System
6.1 Introduction
6.2 Overview of Vehicular Networks
6.2.1 Vehicular Networks Components
6.2.2 Communication Architectures
6.2.2.1 Centralized Architecture: Vehicle-to-Infrastructure Communication
6.2.2.2 Distributed Architecture: Vehicle-to-Vehicle Communication
6.2.2.3 Hybrid Architecture
6.3 V2X Communications
6.4 Security Requirements
6.4.1 Authentication
6.4.1.1 Authentication of the ID
6.4.1.2 Property Authentication
6.4.2 Integrity
6.4.3 Confidentiality
6.4.4 Non-repudiation
6.4.5 Availability
6.4.6 Access Control
6.5 Preliminaries
6.5.1 Encrypt Using Elliptical Curves
6.5.1.1 Exchange of Keys by Elliptical Curves
6.5.1.2 Transmission of Messages
6.5.2 Attribute-Based Signature
6.5.2.1 Computational Assumption
6.5.2.2 Lagrange Interpolation
6.5.2.3 Attribute-Typed Signature
6.6 Proposed Scheme
6.7 Validation and Evaluation
6.7.1 Verification Environment
6.7.1.1 High-Level Protocol Specification Language (HLPSL)
6.7.1.2 Verification Hypotheses
6.7.1.3 Properties to Check
6.7.2 Security Analysis
6.7.3 Performance Evaluation
6.8 Conclusion
References
7 A Novel Unsupervised Learning Method for Intrusion Detection in Software-Defined Networks
7.1 Introduction
7.2 Related Work
7.3 IDS-IF: An Overview
7.4 IDS-IF
7.5 Evaluation of IDS-IF
7.5.1 Experimental Environment
7.5.2 Performance Evaluation
7.6 Conclusion
References
8 Deep Reinforcement Learning Modeling of a V2V Communication-Based Bike Avoidance Protocol for Increased Vehicular Flow
8.1 Introduction
8.2 Related Work
8.3 Overview
8.3.1 DBDA Protocol
8.3.2 Deep Reinforcement Learning
8.4 Markov Decision Process Modeling for DRL-Based DBDA Protocol
8.5 Deep Q-Learning Architecture for DRL-Based DBDA Protocol
8.5.1 Optimal Driving Policy Estimation
8.5.2 Deep Neural Network Architecture for Optimal Driving Learning
8.6 Discussion
8.7 Conclusion
References
9 Deep Learning-Based Modeling of Pedestrian Perception and Decision-Making in Refuge Island for Autonomous Driving
9.1 Introduction
9.2 Related Work
9.3 Overview
9.3.1 APC Protocol
9.3.2 Deep Machine Learning
9.3.2.1 Conventional Neural Network
9.3.2.2 Long Short-Term Memory
9.4 Contribution
9.4.1 P-LPN-Based Architecture for Pedestrian Perception
9.4.2 LSTM Application for Real-Time Decision-Making
9.5 Discussion
9.6 Conclusion
References
10 Machine Learning for Hate Speech Detection in Arabic Social Media
10.1 Introduction
10.2 Overview of Hate Speech Detection on Arabic Social Media
10.3 Natural Language Processing
10.3.1 Stemming
10.3.2 Bag of Words and Term Frequency
10.3.3 Term Frequency-Inverse Document Frequency (TF-IDF)
10.4 Dataset
10.4.1 YouTube
10.4.2 Alakrot's YouTube Comments Collection [10]
10.5 Methodology
10.5.1 Preprocessing
10.5.2 Algorithms
10.5.2.1 Logistic Regression (LR)
10.5.2.2 Random Forests (RF)
10.5.2.3 Support Vector Machines (SVM)
10.5.2.4 Long Short-Term Memory (LSTM)
10.5.3 Evaluation Metrics
10.6 Experimental Results
10.7 Conclusion and Perspectives
References
11 PDDL Planning and Ontologies, a Tool for Automatic Composition of Intentional-Contextual Web Services
11.1 Introduction
11.2 Background and Related Work
11.2.1 Context and Context Awareness
11.2.2 Intention
11.2.3 Service Composition and Related Work
11.2.3.1 Service Composition Categories
11.2.3.2 Technical Classification of Service Composition Approaches
11.2.3.3 Service Composition Approaches
11.2.4 Summary
11.3 Intention and Context Modelling
11.3.1 Proposed Meta-Model
11.3.2 Ontology-Based Intention and Context Modeller
11.3.3 Intention Ontology
11.3.4 Context Ontology
11.3.5 OWL-S Extension for the Semantic Integration of Context and Intention
11.4 PDDL and OWL Interaction
11.4.1 Planning and PDDL
11.4.2 Mapping OWL to PDDL
11.5 Proposed Architecture Overview
11.5.1 CISCA Architecture
11.5.2 CISCA Architecture Features
11.6 Service Composition Module
11.7 A Walk-through Example
11.8 Conclusion and Future Work
References
12 QSAR Anti-HIV Feature Selection and Prediction for Drug Discovery Using Genetic Algorithm and Machine Learning Algorithms
12.1 Introduction
12.2 Proposed Techniques
12.2.1 Genetic Algorithms
12.2.2 Machine Learning Methods in QSAR Problem
12.3 Proposed Approach and Validation
12.3.1 Data
12.3.2 Proposed Approach
12.3.3 Results and Validation
12.4 Conclusion
References
13 Mining Electronic Health Records of Patients Using Linked Data for Ranking Diseases
13.1 Introduction
13.2 Related Work
13.2.1 Mining Electronic Health Records Using Linked Data
13.2.2 Diseases Identification Based on Clustering RDF Dataset
13.3 Results and Discussion
13.3.1 Dataset
13.3.2 Validation Metrics
13.3.3 Quality of Results and Discussion
13.4 Conclusion and Future Work
References
14 The COVID-19 Pandemic's Impact on Stock Markets and Economy: Deep Neural Networks Driving the Alpha Factors Ranking
14.1 Introduction
14.2 Theoretical Background
14.2.1 Investment Factors
14.2.2 Cross-Sectional Investment
14.3 Cross-Sectional Investment-Based Clustering and Intelligent Delay
14.4 The COVID-19 Influence on the Markets and Economy
14.4.1 COVID-19 Growth Analysis
14.4.2 2008 Crisis Comparison
14.4.3 The Influence of COVID-19 on Companies According to their Geographic Revenue
14.4.4 Delay
14.5 Discussion
14.6 Conclusion
Appendix
Original and Customized Factors Utilized in this Work
References
15 An Artificial Immune System for the Management of the Emergency Divisions
15.1 Introduction
15.2 Review of the Literature
15.2.1 Disruptions in Healthcare Facilities
15.2.2 The Status of Stress Within the Emergency Divisions
15.2.3 Improvement Actions to Deal with Situations of Tension
15.2.4 White Plan: The Guiding Standards
15.2.5 Improving the Capacity of Reception Within Hospitals
15.3 Artificial Immune System Techniques: An Overview
15.3.1 Key Conception
15.3.2 Negative Selection
15.3.3 Clonal Selection
15.3.4 Immunity Networks
15.3.5 Training Methods
15.4 Scrutiny of the Arriving Patients
15.5 Prioritizing Patients at Emergency Divisions
15.6 Projection of Filtering Methodology
15.6.1 Overview of the System
15.6.2 Distance Metrics Function
15.6.3 Choosing Manhattan Distance
15.7 Manhattan Distance: The Matching Method
15.7.1 Defining the Problem
15.7.2 Framework of Optimization
15.8 System Skeleton
15.8.1 Gathering Traces
15.8.2 Scrutinizing Traces
15.8.2.1 Basic Concept of Negative Selectivity NSA
15.9 Outcomes of Computation
15.9.1 Alimenting the Database
15.10 Conclusion
References
Index