声明
ABSTRACT
摘要
TABLE OF CONTENTS
LIST OF FIGURES
LIST OF TABLES
CHAPTER 1:INTRODUCTION
1.1 Background
1.2 Cyber-Physical Systems
1.3 Design Challenges
1.4 Problem Definition(Need for CPS Security)
1.5 Cyber-Physieal System Workflow
1.6 Security Requirements
1.7 Challenges in Cyber-Physical Security
1.8 Motivation
1.9 Current Research Situation
1.10 Research Contribution
1.11 Summary
1.12 Thesis Outline
CHAPTER 2:LITERATURE REVIEW
2.1 Introduction
2.2 Related work
2.3 Attack Models in Wireless Sensor and Actuator Networks
2.3.1 Identity Attacks
2.3.2 Route-based Attacks
2.3.3 Network Intrusion
2.4 Miscellaneous Attacks
2.4.1 Physical layer Attacks
2.4.2 MAC layer Attacks
2.4.3 Attacks against the Base Station
2.5 Security Requirements for Wireless Sensor and Actuator Networks
2.6 Potential Threats
2.6.1 Eavesdropping
2.6.2 Spoofing
2.6.3 Denial of Service
2.7 Distributed Denial of Service-The Internet
2.8 Distributed Denial of Service-Wireless Sensor and Actuator Networks
2.9 Security Approaches
2.9.1 Confidentiality
2.9.2 Integrity
2.9.3 Authentication
2.9.4 Authorization
2.10 Summary
CHAPTER 3:PROPOSED CPS ARCHITECTURE
3.1 Introduction:Why is CPS Considered as a Group of WSANs?
3.2 WIRELESS SENSOR AND ACTUATOR NETWORK(WSAN)
3.2.1 Terminology
3.3 ADVANTAGES OF OURPROPOSED CPS ARCHITECTURE
3.4 DDoS Attack Pattern Modeling
3.5 Requirements for DDoS Attack Detection in our proposed CPS architecture
3.6 Adversary Model
3.6.1:Injected sensor nodes
3.6.2:Compromised nodes
3.6.3:Laptop-class nodes
3.7 Network model
3.7.1 Flat Topology
3.7.2 Cluster-based Topology
3.7.3 Data Aggregation Topology
3.8 Threshold Pattern Modeling
3.9 Traffic Flow Observation Table
3.10 Conclusions
CHAPTER 4:DISTRIBUTED ATTACK DETECTION SCHEME
4.1 Introduction
4.1.1 Preliminaries
4.1.2 Contributions
4.2 Attack Detection Scheme
4.2.1 Phase A:Initialization
4.2.2 Phase D:Observation
4.2.3 Phase C:Communication
4.2.4 Phase 4:Verdict
4.2.5 Phase 5:Pattern Update
4.3 Efficiency Analysis
4.4 Conclusions
CHAPTER 5:PERFORMANCE ANALYSIS AND BENCHMARKING
5.1 Introduction
5.2 Analysis
5.2.1 Experimental Setup
5.2.2 Energy Decay Rates
5.2.3 Attack Detection Rates
5.2.4 Pattern Update Rate
5.2.5 False Alarm Rates
5.3 Self-Organizing Map-based Attack Detection
5.3.1 Learning Phase
5.3.2 Data Classification
5.3.3 Parameter selection
5.3.4 Evaluation
5.4 Comparative Analysis
5.5 Compromise-Tolerant Attack Detection Scheme
5.6 Optimal Cluster Size
5.7 Attack Detection Scheme
5.8 Evaluation
5.8.1 Experimental Setup
5.8.2 Simulation Parameters
5.8.3 Analysis
5.9 Conclusions
CONCLUSION
ACKNOWLEDEMENTS
REFERENCES