AI 中文总结
该研究针对复杂网络上的SIS模型网络流行病，通过模拟分析不同网络安全场景，发现针对性隔离关键节点可延迟疫情爆发并降低稳态感染，为网络基础设施防护提供了依据。
AI 中文摘要
高度连接且异构的数字基础设施的快速扩张从根本上改变了网络流行病的动态，使得在现实结构和政策条件下理解恶意软件如何传播变得愈发重要。本研究在复杂网络上采用SIS流行病模型，研究一系列网络安全场景下的病毒传播情况，包括受保护程度最低的环境、杀毒软件、易受攻击的遗留系统、混合安全系统、快速演化的病毒以及关键节点的针对性隔离。通过在无标度拓扑上进行模拟，我们描述了感染流行率随时间的演变情况，并确定了疫情爆发为爆炸性传播或得到控制的条件。结果显示，易受攻击的系统和快速演化的病毒场景会导致快速且几乎完全的感染，而杀毒软件和混合安全配置可显著降低峰值感染水平。最重要的是，我们证明，对一小部分具有结构重要性的节点进行针对性隔离，既可以延迟大规模疫情的爆发，又可以降低稳态感染水平，同时确定了改变计算机病毒传播所需的隔离节点比例的阈值。这些发现强调了网络拓扑和选择性保护在增强网络物理基础设施弹性方面的关键作用。
英文摘要
The rapid expansion of highly connected and heterogeneous digital infrastructures has fundamentally altered the dynamics of cyber epidemics, making it increasingly important to understand how malware propagates under realistic structural and policy conditions. In this study, we employ an SIS epidemic model on complex networks to investigate virus spreading under a range of cybersecurity scenarios, including minimally protected environments, anti-virus, vulnerable legacy systems, mixed security systems, rapidly evolving virus, and targeted quarantine of key nodes. Using simulations on scale-free topologies, we characterize how infection prevalence evolves over time and identify conditions under which outbreaks become explosive versus controlled. Our results show that vulnerable and rapidly evolving virus scenarios lead to fast and almost complete compromise, while antivirus and mixed-security configurations significantly reduce peak infection levels. Most importantly, we demonstrate that targeted quarantine of a small fraction of structurally important nodes can both delay the onset of large scale outbreaks and lower steady state infection, revealing a threshold in the fraction of quarantined nodes which required to alter spreading of computer virus. These findings highlight the critical role of network topology and selective protection in enhancing the resilience of cyber-physical infrastructures.
