Acknowledgements | 第4页 |
Dedication | 第4-5页 |
摘要 | 第5-6页 |
Abstract | 第6-7页 |
List of Abbreviations | 第14-15页 |
1 Introduction | 第15-26页 |
1.1 Background | 第15-16页 |
1.2 APT Attacks in Cloud Computing | 第16-18页 |
1.3 Modeling APT Attacks in Cloud Computing Networks | 第18-19页 |
1.4 Problem Statement and Research Questions | 第19-22页 |
1.5 Innovations and Contributions | 第22页 |
1.6 Scope and Significance of the study | 第22-24页 |
1.7 Organization of the dissertation | 第24-26页 |
2 Literature Review | 第26-45页 |
2.1 Cyber-attacks in Cloud Computing Service Models | 第26-30页 |
2.2 Cyber-attacks on Cloud Computing Deployment Models | 第30-32页 |
2.3 Cyber and APT Attacks Modeling | 第32-43页 |
2.3.1 Cyber-attacks Modeling Approaches | 第38-40页 |
2.3.2 APT Attacks Modeling Approaches | 第40-43页 |
2.4 Literature Review Summary | 第43-45页 |
3 Design of the Modeling Methodology | 第45-55页 |
3.1 Finite State Machines Model-APT States Modeling | 第46页 |
3.2 Bayesian Networks Model-Vulnerability Exploitation | 第46页 |
3.3 Complex Networks Model-Detection Modeling | 第46页 |
3.4 APTs Botnets Utilization | 第46-48页 |
3.5 Datasets | 第48-52页 |
3.5.1 Data Processing and Analysis Methods | 第50页 |
3.5.2 CVEs Datasets Processing | 第50-51页 |
3.5.3 LANL Datasets Processing | 第51-52页 |
3.6 Tools and Hardware Considerations | 第52-55页 |
3.6.1 Data Processing Tools | 第52-53页 |
3.6.2 Data Manipulation and Evaluation Tools | 第53页 |
3.6.3 Network Graphing and Visualization Tools | 第53-54页 |
3.6.4 Data Clustering and Classification Tools | 第54页 |
3.6.5 Hardware and Testbed Environments | 第54页 |
3.6.6 Scope and Limitations | 第54-55页 |
4 The Bayesian Networks APT Attack Model | 第55-72页 |
4.1 APT Attackers Profiling | 第55-56页 |
4.2 Attacker's perception vs Actual system exploitability | 第56-58页 |
4.3 Cloud Infrastructure Layer Partitioning | 第58-72页 |
4.3.1 Attack Paths Formalizations | 第59-63页 |
4.3.2 The Bayesian Attack Network | 第63-64页 |
4.3.3 Conditional Probabilities with detection | 第64-65页 |
4.3.4 Path Derivations and Conditional Probability Assignments | 第65-67页 |
4.3.5 Optimized Shortest Path Algorithm and Edge Weighting | 第67-69页 |
4.3.6 Attack Complexity and Time Cost | 第69-72页 |
5 Finite State Machine Model for APT Attacks | 第72-88页 |
5.1 FSM Model for APT Attacks on Discrete Hosts | 第72-81页 |
5.1.1 Security States and Transitions of a Discrete Host | 第73-79页 |
5.1.2 Formulation of the APT Attack Model | 第79页 |
5.1.3 Attack Tree Integration and Analysis | 第79-81页 |
5.2 FSM Model for Bayesian Networks APT Attacks | 第81-84页 |
5.2.1 APT Attack Source | 第82-83页 |
5.2.2 APT Attack State | 第83页 |
5.2.3 APT Attack Nodes | 第83-84页 |
5.3 Global FSM Model for APT Attacks | 第84-88页 |
5.3.1 APT Attack State Transition Table | 第85-86页 |
5.3.2 APT Attacks K-maps | 第86-88页 |
6 Complex Networks Model for APT Attacks Detection | 第88-102页 |
6.1 Unpredictability of APT Attack Lifecycle Stages | 第88-90页 |
6.2 Dynamism of APT-ANs and Communication Networks | 第90-92页 |
6.3 Imbalanced Data Distribution | 第92-94页 |
6.4 Small World Communication Network Model | 第94-95页 |
6.5 Scale-Free APT-AN Network Model | 第95-99页 |
6.6 Scarcity of Public APT Data | 第99页 |
6.7 FSM State Changes of Complex Network Nodes in APT-ANs | 第99-102页 |
7 Data Preprocessing and Formatting | 第102-111页 |
7.1 CVEs and Base Scores | 第102-105页 |
7.2 Network flows and DNS | 第105-109页 |
7.3 Feature Normalization | 第109-111页 |
8 Modeling Results and Analyses | 第111-127页 |
8.1 Bayesian Network Based APT Attack Paths | 第111-116页 |
8.2 Detection of Multi-stages APTs by a Semi-supervised LearningApproach | 第116-127页 |
8.2.1 Detection in the Infiltration Phase | 第118-119页 |
8.2.2 Detection in the Lateral Movement Phase | 第119-120页 |
8.2.3 Detection in the C2 Beaconing and Exfiltration Phase | 第120页 |
8.2.4 Application of the Clustering and Classification Algorithms | 第120-127页 |
9 Conclusion and Future Directions | 第127-130页 |
9.1 Conclusion and Significance | 第127-128页 |
9.2 Future Directions | 第128-130页 |
References | 第130-146页 |
作者简历及在学研究成果 | 第146-150页 |
学位论文数据集 | 第150-151页 |