Developing networks using artificial intelligence / Haipeng Yao, Chunxiao Jiang, Yi Qian.

Intro; Preface; Acknowledgments; Contents; 1 Introduction; 1.1 Background; 1.2 Overview of SDN and Machine Learning; 1.2.1 Software Defined Networking (SDN); 1.2.2 Machine Learning; 1.2.2.1 Supervised Learning; 1.2.2.2 Unsupervised Learning; 1.2.2.3 Reinforcement Learning; 1.3 Related Research and Development; 1.3.1 3GPP SA2; 1.3.2 ETSI ISG ENI; 1.3.3 ITU-T FG-ML5G; 1.4 Organizations of This Book; 1.5 Summary; 2 Intelligence-Driven Networking Architecture; 2.1 Network AI: An Intelligent Network Architecture for Self-Learning Control Strategies in Software Defined Networks

2.1.1 Network Architecture2.1.1.1 Forwarding Plane; 2.1.1.2 Control Plane; 2.1.1.3 AI Plane; 2.1.2 Network Control Loop; 2.1.2.1 Action Issue; 2.1.2.2 Network State Upload; 2.1.2.3 Policy Generation; 2.1.3 Use Case; 2.1.4 Challenges and Discussions; 2.1.4.1 Communication Overhead; 2.1.4.2 Training Cost; 2.1.4.3 Testbeds; 2.2 Summary; References; 3 Intelligent Network Awareness; 3.1 Intrusion Detection System Based on Multi-Level Semi-Supervised Machine Learning; 3.1.1 Proposed Scheme (MSML); 3.1.1.1 Pure Cluster Extraction (PCE); 3.1.1.2 Pattern Discovery (PD)

3.1.1.3 Fine-Grained Classification (FC)3.1.1.4 Model Updating; 3.1.1.5 The Hyper-Parameters; 3.1.2 Evaluation; 3.1.2.1 Dataset; 3.1.2.2 Data Pre-process; 3.1.2.3 Evaluation Criteria; 3.1.2.4 Baseline Model; 3.1.2.5 MSML; 3.2 Intrusion Detection Based on Hybrid Multi-Level Data Mining; 3.2.1 The Framework of HMLD; 3.2.2 HMLD with KDDCUP99; 3.2.2.1 KDDCUP99 Dataset; 3.2.2.2 MH-DE Module; 3.2.2.3 MH-ML Module; 3.2.2.4 MEM Module; 3.2.3 Experimental Results and Discussions; 3.2.3.1 Evaluation Criteria; 3.2.3.2 Experiments and Analysis

3.3 Abnormal Network Traffic Detection Based on Big Data Analysis3.3.1 System Model; 3.3.1.1 Normal Traffic Selection Model; 3.3.1.2 Abnormal Traffic Selection Model; 3.3.1.3 Abnormal Traffic Selection Model; 3.3.2 Simulation Results and Discussions; 3.3.2.1 Data Set; 3.3.2.2 Simulation Results; 3.3.2.3 Discussing Result of No. 8 and No. 11; 3.3.2.4 Discussing Result of No. 5 and No. 7; 3.3.2.5 Discussing Result of No. 3 and No. 4; 3.4 Summary; References; 4 Intelligent Network Control; 4.1 Multi-Controller Optimization in SDN; 4.1.1 System Model; 4.1.1.1 Network Model

4.1.1.2 Communication Model4.1.1.3 Computation Model; 4.1.1.4 Problem Formulation; 4.1.2 Methodology; 4.1.2.1 PSO Aided Near-Optimal Multi-Controller Placement; 4.1.2.2 Resource Management Relying on Deep Q-Learning; 4.1.3 Simulation Results; 4.2 QoS-Enabled Load Scheduling Based on ReinforcementLearning; 4.2.1 System Description; 4.2.1.1 Energy Internet; 4.2.1.2 Software-Defined Energy Internet; 4.2.1.3 Controller Mind framework; 4.2.1.4 Re-Queuing Module; 4.2.1.5 Info-Table Module; 4.2.1.6 Learning Module; 4.2.2 System Model; 4.2.2.1 Re-Queuing Model; 4.2.2.2 Workload Model