An approach to generate damage strategies for inter-domain routing systems based on multi-objective optimization

Area of Science:

• Computer Science
• Network Security
• Optimization Theory

Background:

• Inter-domain routing systems are critical internet infrastructure prone to disruptions.
• Previous research on damage strategies often overlooked attack costs and optimization effectiveness.
• Understanding attacker behavior is key to developing robust defense mechanisms.

Purpose of the Study:

• To design an algorithm for generating effective damage strategies for inter-domain routing systems.
• To address limitations in existing research regarding attack cost definition and optimization.
• To improve the selection of optimal attack node groups.

Main Methods:

• Developed a multi-objective optimization algorithm (PMT) for damage strategy generation.
• Transformed the problem into a double-objective optimization task.
• Defined attack cost based on the degree of nonlinearity.
• Proposed network partitioning for initialization and partition search for node replacement.

Main Results:

• The PMT algorithm demonstrated superior effectiveness and accuracy compared to five existing algorithms.
• Experimental results validated the proposed approach for generating damage strategies.
• The new method provides a more realistic consideration of attack costs.

Conclusions:

• The PMT algorithm offers an effective solution for optimizing attack node group selection in inter-domain routing systems.
• The study highlights the importance of incorporating attack costs into damage strategy development.
• PMT provides a more accurate and efficient method for network security analysis.