Machine Learning for Networking: 5th International Conference, MLN 2022, Paris, France, November 28–30, 2022, Revised Selected Papers (Lecture Notes in Computer Science)

Preface
Organization
Contents
Comparison of AI-Based Algorithms for Low Energy Communication
	1 Introduction
	2 Background
		2.1 Standard 802.15.1: Bluetooth Et Bluetooth Low Energy
		2.2 Limiting Factors in a Receiver (analogue Part)
		2.3 Evaluate the Quality of the Link
	3 Usual Countermeasures
	4 AI-Based Countermesures for Cognitive Radio
		4.1 Spectrum Prediction
		4.2 Waveform Classification of Interfering Technologies
		4.3 Temporal Location of Intra-package Damage
	5 Futur Work
	6 Conclusion
	References
Development of an Intent-Based Network Incorporating Machine Learning for Service Assurance of E-Commerce Online Stores
	1 Introduction
	2 Related Work
	3 Proposed IBN Architecture and Machine Learning Method
		3.1 IBN Architecture
		3.2 Machine Learning Method Applied to the IBN Network
	4 Design of Experiments of Machine Learning
	5 Results
	6 Conclusions
	References
Cyber-attack Proactive Defense Using Multivariate Time Series and Machine Learning with Fuzzy Inference-based Decision System
	1 Introduction
	2 Related Work
	3 Background
		3.1 Multi-variate Time Series
		3.2 Fuzzy Inference System
	4 Proposed Approach
		4.1 Data Gathering
		4.2 Multivariate Forecasting
		4.3 Fuzzy Inference-based Detection System (FIDS)
	5 Implementation and Evaluation Results
		5.1 Data Gathering and Forecasting Results
		5.2 FIDS Results
	6 Conclusion and Future Work
	References
iPerfOPS: A Tool for Machine Learning-Based Optimization Through Protocol Selection
	1 Introduction
	2 Background and Related Work
		2.1 Dedicated Vs. Shared Networks: Impact of Background Traffic
		2.2 TCP Scheme: Congestion Control Algorithm (CCA)
		2.3 OPS: Optimization Through Protocol Selection
	3 Methodology
		3.1 ML Process
		3.2 Network Testbed
		3.3 Libraries and Software Configuration
	4 iPerfOPS: An OPS Data-Transfer Tool
		4.1 Development Strategy: Adopting iPerf Library
		4.2 Extending iPerf for Data Transfer
		4.3 iPerfOPS Architecture
	5 Performance Evaluation
	6 Concluding Remarks
	References
GRAPHSEC – Advancing the Application of AI/ML to Network Security Through Graph Neural Networks
	1 AI4SEC and the Present State of Affairs
		1.1 AI4SEC – A Slow(ed) Path to Success
		1.2 AI4SEC and Graph Machine Learning
		1.3 The GRAPHSEC Vision – GNNs for Network Security
	2 AI4SEC and GNN4SEC – State of the Art
		2.1 AI4SEC – AI/ML for Network Security
		2.2 GNNs for NETSEC
		2.3 XAI – Explainable AI for Network Security
		2.4 Data Generation – Data Limitation and Lack of Labels
	3 The Path to GRAPHSEC
		3.1 Research Questions and Hypotheses
		3.2 Expected Results from GRAPHSEC
	4 Conclusions
	References
Low Complexity Adaptive ML Approaches for End-to-End Latency Prediction
	1 Introduction
	2 Related work
	3 Simple Machine-Learning Approaches for Latency Prediction
		3.1 Feature Engineering and Machine Learning
		3.2 Curve Regression by Bernstein Polynomials
	4 Adaptive Versions
	5 Experiments and Results
		5.1 Dataset
		5.2 Results
	6 Conclusion
	References
TDMA-Based MAC Protocols Designed or Optimized Using Artificial Intelligence for Safety Data Dissemination in Vehicular Ad-Hoc Network: A Survey
	1 Introduction and Background
		1.1 Vehicular Communications Types
		1.2 Safety Messages Types in Facilities/Application Layer
		1.3 Motivations
		1.4 Contributions of This Paper
	2 Related Work
		2.1 Classification of MAC Protocols in VANETs
		2.2 Classification of VANET TDMA-Based MAC Protocols on Channel Access Mechanisms
	3 Classification of Artificial Intelligent-Based MAC Protocols
		3.1 Contention-Based MAC Protocols Optimizing Using AI
		3.2 Classification of TDMA-Based Contention-Free MAC Protocols Designed or Optimized Using AI
	4 Artificial Intelligent MAC Protocols Optimizing Methods
		4.1 Classical Machine Learning Algorithms
		4.2 Deep Learning
		4.3 Reinforcement Learning
		4.4 Heuristic Algorithms
	5 Real Dataset
		5.1 Highway Roads
		5.2 Urban Roads
	6 Conclusion
	References
A Machine Learning Based Approach to Detect Stealthy Cobalt Strike C&C Activities from Encrypted Network Traffic
	1 Introduction
	2 Analysis of Cobalt Strike C&C
		2.1 Cobalt Strike
		2.2 Cobalt Strike C&C Communications
		2.3 Malleable C&C Profiles
	3 A Machine Learning Based Detection
		3.1 Threat Model
		3.2 Flow Based Features
		3.3 Machine Learning Models
		3.4 Evaluation Metrics
	4 Empirical Validation
		4.1 Dataset Acquisition
		4.2 Flow Based Machine Learning Detection
	5 Related Works
	6 Conclusions
	References
Unified Emulation-Simulation Training Environment for Autonomous Cyber Agents
	1 Introduction
	2 Unified CyGIL-E and CyGIL-S
		2.1 System Design
		2.2 Unsupervised Auto-generation of CyGIL-S
	3 CYGIL-S Evaluation
		3.1 Test Data Collection Time
		3.2 Unknown Transitions in CyGIL-S
	4 Unified CyGIL Training
		4.1 The Cross Training Loop
		4.2 Discussions
	5 Concluding Remarks
	References
Deep Learning Based Camera Switching for Sports Broadcasting
	1 Introduction
	2 Our Approach
		2.1 Data Collection and Labeling
		2.2 Our Deep Learning Network
		2.3 Camera Switching
	3 Evaluation and Discussion
	4 Conclusions
	References
Phisherman: Phishing Link Scanner
	1 Introduction
		1.1 Problems Encountered/Opportunities
		1.2 Research Objective
		1.3 Scope and Limitations
		1.4 Significance of the Study
	2 Research Methodology
		2.1 Planning
		2.2 Design and Development
		2.3 Deployment
	3 Results and Discussion
	4 Conclusion and Recommendation
		4.1 Conclusion
		4.2 Recommendation
	References
Leader-Assisted Client Selection for Federated Learning in IoT via the Cooperation of Nearby Devices
	1 Introduction
	2 Related Work
	3 Leader-Assisted Client Selection Approach
		3.1 Leader Election
		3.2 Gathering Process
		3.3 Selection Using Lightweight Deep Learning
		3.4 Electing a New Leader
	4 Experiments
		4.1 Simulation Setup
		4.2 Results
	5 Conclusion
	References
Author Index