Advances in Nature-Inspired Cyber Security and Resilience

Preface
Contents
About the Editors
Nature-Inspired Cybersecurity and Resilience: An Overview
1 Introduction
2 Literature Survey
3 Noteworthy Contributions
4 Conclusion
References
Detection of Reconnaissance Attacks on IoT Devices Using Deep Neural Networks
1 Introduction
1.1 IoT
1.2 Reconnaissance Attacks
1.3 Reconnaissance Attacks in IoT
1.4 Problem Statement
2 Related Works
2.1 Classical Detection of Reconnaissance Attacks
2.2 Machine-Learning Based Detection of Reconnaissance Attacks
3 Methodology
3.1 Dataset Selection
3.2 Feature Extraction
3.3 Dataset Preprocessing
3.4 Deep Neural Network Architecture
4 Implementation
4.1 Implementation Environment
4.2 Training
4.2.1 Initial Training
4.2.2 10-Fold Cross-Validation
5 Results
6 Discussions
7 Conclusions and Future Work
References
Particle Swarm Optimization-Driven DSE-Based Low-Cost Hardware Security for Securing DSP IP Cores
1 Introduction
2 What Role a Nature-Inspired Algorithm Can Play in Hardware Security
2.1 Overview of PSO-Based Nature-Inspired Algorithm for DSE [10, 13]
3 Low-Cost Hardware Security Techniques Integrated with PSO-DSE
3.1 Low-Cost Hardware Trojan Security Technique [12]
3.1.1 Threat Model
3.1.2 Hardware Trojans in 3PIP Blocks/Modules [12]
3.1.3 Exploring Low-Cost Trojan Secured DMR Datapath Using PSO-Based DSE [12]
3.2 Low-Cost Hardware Watermarking Technique [13]
3.2.1 Threat Model
3.2.2 Multivariable Watermarking [13]
3.2.3 Exploring Low-Cost Watermarked Solution Using PSO-Based DSE [13]
4 Conclusion
References
Malicious Activity Detection in IoT Networks: A Nature-Inspired Approach
1 Introduction
2 Vulnerabilities in IoT Networks
3 Cyber Attacks in IoT Networks
4 Intrusion Detection Systems
5 Evolutionary Computing Algorithms
5.1 Evolutionary Computing Basic Concepts
5.1.1 Populations
5.1.2 Selection
5.1.3 Crossover
5.1.4 Mutation
5.2 Evolutionary Computing Algorithmic Steps
5.3 Genetic Algorithm
5.4 Genetic Programming
5.5 Evolutionary Strategy
5.6 Evolutionary Programming
5.7 Evolution Strategies Applied in Intrusion Detection
6 Artificial Immune Systems
6.1 Artificial Immune Systems Applied in Intrusion Detection
6.2 Fuzzy Logic
6.3 Fuzzy Logic Applied in Intrusion Detection
6.4 Chaos Theory
6.5 Chaos Theory Applied in Intrusion Detection
7 Swarm Intelligence Algorithms
7.1 Swarm Intelligence Fundamentals
7.1.1 Proximity
7.1.2 Quality
7.1.3 Diverse Response Methods
7.1.4 Stability
7.1.5 Adaptability
7.2 Particle Swarm Optimisation Algorithm
7.3 Ant Colony Optimisation Algorithm
7.4 Artificial Bee Colony Algorithm
7.5 Fish Swarm Algorithm
7.6 Firefly Algorithm
7.7 Swarm Intelligence Applied in Intrusion Detection
8 Artificial Neural Networks
8.1 Artificial Neural Networks Components
8.1.1 Neurons
8.1.2 Connections and Weights
8.1.3 Activation Function
8.2 Artificial Neural Networks Training Procedure
8.2.1 Hidden Layers Number
8.2.2 Learning Rate
8.2.3 Momentum
8.2.4 Cost Function
8.2.5 Epochs Number and Batch Size
8.3 Popular Artificial Neural Network Architectures
8.3.1 Feedforward Neural Networks
8.3.2 Recurrent Neural Networks
8.4 Artificial Neural Networks Applied in Intrusion Detection
9 Discussion and Future Directions
10 Conclusions
References
Nature-Inspired Malware and Anomaly Detection in Android-Based Systems
1 Introduction
2 Cybersecurity Criterion with Nature-Inspired Solutions
2.1 Common Expressions in Bio-inspired Computing and NICS
2.1.1 Population
2.1.2 Generation
2.1.3 Mutation
2.1.4 Crossover
2.1.5 Fitness and Selection
2.2 Existing Nature-Inspired Algorithms
2.2.1 Inspired from Mammalian Body
2.2.2 Inspired by Swarms of Organisms
2.2.3 Inspired by the Evolution of Organisms
2.2.4 Chaos Theory
2.3 Nature-Inspired Approach
2.3.1 Passive Approach
2.3.2 Active Approach
2.4 State of Nature-Inspired Solutions
3 Features of Nature-Inspired Solutions
3.1 Modular Design
3.2 External Rules and Constraints
3.3 Internal/Central Behavioural Rules and Constraints
3.4 Inter-Independence and Emergent Behaviours
3.5 Indiscriminate Disturbances in Environment and Input
3.6 Dynamic Circumstances and Restrictions
4 Anomaly and Malware Detection for Android-Based Systems
4.1 Android Malware and Anomaly
4.1.1 Android Operating System
4.1.2 Malware
4.1.3 Anomaly
4.1.4 Features of Android Malware and Anomalies
4.2 Existing Generic Malware Detection Methods
4.2.1 Signature-Based Detection
4.2.2 Behavioural Detection
4.2.3 Machine Learning-Based Detection
4.3 Contemporary Issues in Malware Detection
5 Nature-Inspired Malware and Anomaly Detection
5.1 Nature-Inspired Android Malware Detection
5.1.1 Passive NICS for Android Malware
5.1.2 Active NICS for Android Malware
6 Application and Adaptability of NICS Solutions
7 Research Challenges and Shortcomings in NICS
8 Conclusion
References
A Review of Nature-Inspired Artificial Intelligence and Machine Learning Methods for Cybersecurity Applications
1 Introduction
1.1 Background and Motivation
2 Artificial Intelligence and Machine Learning in Cybersecurity
2.1 Artificial Intelligence Types and Applications
2.2 Use of Artificial Intelligence in Cybersecurity
2.3 Potential Applications of Artificial Intelligence in Cybersecurity
2.4 Application of Deep Learning Techniques for Malware and Intrusion Detection
3 Nature-Inspired Deep Neural Network-Based Learning for Cyber Applications
4 Conclusion
References
A Nature-Inspired DNA Encoding Technique for Quantum Session Key Exchange Protocol
1 Introduction
2 Related Work
3 The DNA Code Set
4 Key Exchange Protocol
5 Encryption and Decryption of the Plain Text: The Quantum Session Key Exchange Protocol (QSKEP)
5.1 Plain Text to DNA Encoded Text and Vice Versa
5.2 Encryption-Decryption of the Message Using Quantum Protocol
5.3 Illustration
6 Security Analysis of QSKEP
7 Error Analysis
8 Conclusion
References
Novel Hybridized Crow Optimization for Secure Data Transmission in Cyber Networks
1 Introduction
1.1 Research Motivation and Contributions
2 Related Works
2.1 Crow Search Algorithm for Clustering in WSNs
2.2 Quality of Service (QoS) in Wireless Sensor Network (WSN)
3 Proposed Methodology
3.1 QOS-CR
3.1.1 Calculation of Trust Factor to Choose the Secure Nodes
3.1.2 Computation of Trust
3.1.3 Computation of Direct Trust
3.1.4 Computation of Indirect Trust
3.1.5 Computation of Recent Trust
3.1.6 Trust that Rely on Data Bytes
3.1.7 Energy Model of Network
3.2 Novel Hybrid Crow Optimization
3.3 QoS-Based Bipartite Neighborhood Clustering Algorithm (QOS-CR)
4 Results and Discussion
5 Conclusion
References
Malware Attacks: Dimensions, Impact, and Defenses
1 Introduction
2 Statistics Related to Malware
2.1 Malware Growth
2.2 Financial Loss Caused by Malware
3 Launch of the Attack
3.1 Generation of Malware
3.1.1 The First Generation Malware (1966–1988): The Era of Virus
3.1.2 The Second Generation Malware (1989–2004): The Era of Worm
3.1.3 The Third Generation (2004–2010): The Era of Bot
3.1.4 The Fourth Generation (2010 Onwards): Era of Cyber War
3.2 Stages of Malware Attack
3.2.1 Malware Writing
3.2.2 Propagation
3.2.3 Payload Execution
4 Defense Against Malware
4.1 Malware Detection
4.1.1 Signature-Based Detection
4.1.2 Non-Signature-Based Detection
4.1.3 Bio-Inspired Defense Against Malware
5 Recent Trends in Malware
5.1 Ransomware
5.2 Document-Based Malware
5.3 Fileless Malware
6 Research Direction
6.1 New Horizons of Malware Attacks and Defense
6.1.1 Malware in Internet of Things (IoT)
6.1.2 Malware in Industrial Control Systems (ICS)
6.1.3 Android Malware
6.1.4 Malicious Hardware
6.2 Existing Issues in Malware Detection
6.2.1 Addressing Issues of Signature-Based Detection
6.2.2 Improving Performance of Non-Signature-Based Techniques
6.2.3 Cleaning and Repairing Devices After Malware Infection
7 Conclusion
References
Index