Onion-like networks are both robust and resilient

Area of Science:

• Network Science
• Complex Systems
• Cybersecurity

Background:

• Network connectivity and flow transmission are vital but threatened by intelligent attacks and cascading failures.
• Existing methods to enhance network robustness, like interwoven loops, are computationally intractable.
• Scale-free networks, common in real systems, exhibit vulnerabilities to such disruptions.

Purpose of the Study:

• To propose a novel method for constructing robust networks with enhanced tolerance to attacks.
• To investigate the resilience of these new networks against cascading overload failures.
• To compare the protective capabilities of the proposed network structure against established models.

Main Methods:

• Constructing onion-like networks through incremental growth, enhancing interwoven loops.
• Introducing positive degree-degree correlations in the network topology.
• Implementing adaptive routing policies for flow control to manage cascading failures.

Main Results:

• The proposed onion-like networks demonstrate optimal tolerance to connectivity attacks, outperforming existing state-of-the-art.
• These networks exhibit adaptive resilience, effectively absorbing failures from single and multiple attacks.
• The inhibitory effect on cascading failures in onion-like networks is significantly stronger than in scale-free networks.

Conclusions:

• Onion-like networks offer a computationally tractable and highly effective approach to enhancing network robustness.
• Adaptive routing policies further bolster the resilience of these networks against cascading overload failures.
• This novel network architecture presents a promising solution for critical infrastructure protection and reliable network function.