Multi-Agent for Network Security Monitoring and Warning: A Generative AI Solution
Qingjun Yuan, Qianwei Meng, Jing Tao, Guangsong Li, Jinlong Fei, Bin Lu, Yongjuan Wang
Abstract
With the development of internet technology, intelligent and covert threat methods, such as zero-day attacks, Advanced Persistent Threats, and ransomware attacks, have emerged in an endless stream, making the network security landscape increasingly complex. In this context, network security monitoring and warning systems, as a critical defense line to ensure cyberspace security, are of paramount importance. This paper analyzes the history and current status of traditional security monitoring systems, highlighting unresolved research issues such as insufficient capabilities and low efficiency. Thanks to the rapid development of Generative Artificial Intelligence, network security monitoring and warning are transitioning from traditional models to automated and intelligent systems, achieving significant results. To effectively address these challenges, there is an urgent need to introduce more advanced technological approaches, based on multi-agent systems that simulate human security expert collaboration, to build a next-generation network security monitoring system that combines accuracy, efficiency, and intelligence, thereby enhancing network security monitoring and warning capabilities comprehensively. The DeepSeek-V3 model was utilized for language refinement in this paper<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup>.