Robustness Optimization for IoT Topology

✍ Scribed by Tie Qiu, Ning Chen, Songwei Zhang

• Publisher: Springer
• Year: 2022
• Tongue: English
• Leaves: 224
• Edition: 1st ed. 2022
• Category: Library

✦ Synopsis

The IoT topology defines the way various components communicate with each other within a network. Topologies can vary greatly in terms of security, power consumption, cost, and complexity. Optimizing the IoT topology for different applications and requirements can help to boost the network’s performance and save costs. More importantly, optimizing the topology robustness can ensure security and prevent network failure at the foundation level. In this context, this book examines the optimization schemes for topology robustness in the IoT, helping readers to construct a robustness optimization framework, from self-organizing to intelligent networking.

The book provides the relevant theoretical framework and the latest empirical research on robustness optimization of IoT topology. Starting with the self-organization of networks, it gradually moves to genetic evolution. It also discusses the application of neural networks and reinforcement learning to endow the node with self-learning ability to allow intelligent networking.

This book is intended for students, practitioners, industry professionals, and researchers who are eager to comprehend the vulnerabilities of IoT topology. It helps them to master the research framework for IoT topology robustness optimization and to build more efficient and reliable IoT topologies in their industry.

✦ Table of Contents

Preface
Contents
Acronyms
1 Introduction
1.1 Background and Motivation
1.2 Characteristics of IoT Topology
1.2.1 Small-World Model in IoT Topology
1.2.2 Scale-Free Model in IoT Topology
1.3 Attack Strategies Against Network Topology
1.3.1 Existing Attack Strategies
1.3.2 Different Centralities and Network Topologies
1.3.3 Attack Strategy of Largest Component
1.4 Book Organization
References
2 Preliminaries of Robustness Optimization
2.1 Metrics of Topology Robustness
2.1.1 Algebraic Connectivity and Natural Connectivity
2.1.2 Metric R and Cumulative k-MVC-Impact
2.1.3 Temporal Robustness and Metrics Based on Betweenness
2.1.4 Three Robustness Metrics of Power Grids
2.2 Related Work
2.2.1 Robustness Optimization Based on Small World Model
2.2.1.1 Wireless Networks
2.2.1.2 Financial Networks
2.2.1.3 Power Grid
2.2.1.4 Transportation Networks
2.2.1.5 P2P Networks
2.2.2 Robustness Optimization Based on Scale-Free Model
2.2.2.1 Onion-Like Structure Resistant to Malicious Attacks
2.2.2.2 Robustness Optimization Strategies for Scale-Free Networks
References
3 Robustness Optimization Based on Self-Organization
3.1 Path Planning with The Greedy Principles
3.1.1 Overview of Topological Path Planning
3.1.2 Description of Greedy Principles
3.1.3 Topological Staining Operation
3.1.4 Shortcuts Addition Strategy Based on Local Importance
3.1.5 Performance Evaluation
3.2 Construction of Highly Robust Topology
3.2.1 Research Background
3.2.2 Preliminaries
3.2.3 Scale-Free Topology Deployment Strategy
3.2.4 Highly Robust Topology Optimization Strategy
3.2.5 Performance Evaluation
3.3 Robust Time Synchronization Scheme
3.3.1 Research Background
3.3.2 Motivation and Preliminaries
3.3.3 Details of Time Synchronization Scheme
3.3.4 Performance Evaluation
References
4 Robustness Optimization Based on Genetic Evolution
4.1 Introduction
4.2 Robustness Optimization Scheme with Multi-population Co-evolution
4.2.1 Optimization Problem Model
4.2.2 Topological Data Processing
4.2.3 Initialization Operation
4.2.4 Crossover Operator
4.2.5 Mutation Operator
4.2.6 Migration Operator
4.2.7 Performance Evaluation
4.3 An Adaptive Robustness Evolution Algorithm with Self-competition
4.3.1 Population Diversity Measurement Method
4.3.1.1 Gene Position Coverage Ratio
4.3.1.2 Gene Distribution Uniformity
4.3.2 Adaptive Adjustment
4.3.3 Self-competitive Mechanism
4.3.4 Performance Evaluation
References
5 Robustness Optimization Based on Swarm Intelligence
5.1 Topology Optimization Strategy with Ant Colony Algorithm
5.1.1 Statement of Research Problem
5.1.2 Preliminaries
5.1.3 Path Search Process Based on Ant Colony
5.1.4 Shortcuts Addition Strategy Based on Global Importance
5.1.5 Performance Evaluation
5.2 Topology Optimization Strategy with Particle Swarm Algorithm
5.2.1 Single Sink and Multiple Sink Networks
5.2.2 Preliminaries
5.2.3 Particle Encoding and Fitness Function
5.2.4 Particle Update Strategy
5.2.5 Performance Evaluation
References
6 Robustness Optimization Based on Multi-Objective Cooperation
6.1 Statement of Problem
6.2 Network Topology Initialization
6.3 Multi-Objective Functions
6.3.1 Energy Efficiency on Network
6.3.2 Load Balancing on Multiple sink nodes
6.3.3 Fitness Function
6.4 Details of Algorithm
6.4.1 Topology Encoding
6.4.2 Crossover and Mutation
6.4.3 Pareto Layering
6.4.4 Layered-Cooperation Mechanism
6.5 Performance Evaluation
References
7 Robustness Optimization Based on Self-Learning
7.1 Malicious Node Identification Scheme Based on Gaussian Mixture Model
7.1.1 Statement of Problem
7.1.2 Data Preprocessing Operations
7.1.3 Parameter Estimation of the Model
7.1.4 Training and Prediction Process of the Model
7.1.5 Performance Evaluation
7.2 Highly Robust Topology Learning Model Based on Neural Network
7.2.1 Learning Framework based on Neural Network
7.2.2 Preliminary Details
7.2.3 Performance Evaluation
7.3 Highly Robust Topology Generation Strategy Based on Temporal Convolutional Network
7.3.1 Statement of Problem
7.3.2 Feasibility Analysis of Sequence Prediction
7.3.3 Construction of Topological Sample Sequence Model
7.3.4 Model Learning and Topology Reconfiguration Strategies
7.3.5 Performance Evaluation
References
8 Robustness Optimization Based on Node Self-Learning
8.1 Problem Model
8.1.1 Topology State
8.1.2 Actor—Critic Model
8.1.3 Action Mapping
8.1.4 Exploration and Exploitation
8.1.5 Algorithm Design
8.2 Performance Evaluation
8.2.1 Convergence of DDLP
8.2.2 Robustness Optimization of IoT Topology
8.2.3 Comparison Between Initial and Optimized Topologies
8.2.4 Comparison Between the Algorithm and Other Algorithms in Different Link Density
References
9 Future Research Directions
9.1 Theory Exploration for Future Networks
9.1.1 Real-Time Topology Robustness Optimization for Ultra-Dense Networks
9.1.2 The Influence of Network Dynamic Local Motif on Global Network
9.1.2.1 Dynamic Evolution of Ultra-Dense Network Topology
9.1.2.2 Which Type of Network Local Motif Has the Greatest Importance in Global Network Topology
9.1.3 A Dynamic and Comprehensive Measure of Network Robustness
9.1.4 Quantum Computing to Improve the Robustness Optimization Speed for Ultra-Dense Networks
9.2 Industrial Applications
9.2.1 Industrial Internet of Things
9.2.2 Smart City
9.2.3 Smart Underwater Internet of Things
References