Load balancing optimization for network anti-attack based on vital node splitting

¹ School of Electronics and Information, Zhengzhou University of Aeronautics - Zhengzhou, China
² Henan Key Laboratory of General Aviation Technology, Zhengzhou University of Aeronautics - Zhengzhou, China
³ Henan Province Collaborative Innovation Center of Aeronautics and Astronautics Electronic Information Technology, Zhengzhou University of Aeronautics - Zhengzhou, China
⁴ Department of Technology, Zhengzhou University of Aeronautics - Zhengzhou, China

Received: 27 February 2025
Accepted: 20 October 2025

Abstract

The Internet, being a heterogeneous scale-free network, is highly vulnerable to selective attacks, making it essential to protect critical nodes for better anti-attack capability. This paper tackles two main challenges: identifying important nodes and improving communication performance after adopting anti-attack strategies. To identify key nodes, the Node Estranging Method (NEM) is proposed, which evaluates node importance by estranging associated edges and defining a measure that considers both global network position and local connection characteristics. Simulations demonstrate that NEM accurately assesses node importance and reduces computational complexity. For improving communication performance, a mathematical model is developed to focus on balancing the communication load of split nodes. An approximate algorithm based on the shortest spanning tree is introduced, yielding results similar to exhaustive methods. The proposed techniques are shown to enhance network robustness and efficiency, providing valuable guidance for network topology optimization and attack mitigation strategies.