基于边权表示学习的乌克兰互联网变化感知

2022年2月至7月期间,网络攻防频繁发生,乌克兰部分地区间接受到断网影响,其境内和边界网络连通情况发生了改变,通过对时间变化拐点的感知和变化区域的分析可敏锐捕捉网络异常变化的细节,为网络威胁预警、网络关键基础设施风险评估等提供重要的信息支持。本文针对2022年2月至7月期间乌克兰自治系统的路由变化进行研究,构建乌克兰自治系统动态网络拓扑,介绍动态网络分析方法,监测乌克兰网络变化,分析断网的时间、地点,并感知断网后网络发生异常的地区。通过与现实事件的对比分析,映射出互联网结构变化感知结果与实体军事冲突的相关性,为从互联网空间观测复杂战场态势提供了信息支撑。

Perception of Internet Changes in Ukraine Based on Edge Weight Representation Learning

From February to July 2022, network attack and defence occurred frequently. As a result of active internet outages and attacks by international hacker organizations outside Ukraine, some regions of Ukraine have been affected, and the situation of internet connectivity within the country and across their borders has changed. Through the analysis of the perception of the inflection point and the change area, the details of the abnormal change of the network can be captured keenly. It provides important information support for network threat early warning and network critical infrastructure research. The changes of the autonomous system topology in Ukraine during this period were studied. The dynamic network topology of Ukraine??s autonomous system was constructed, on the basis of which changes in the Ukrainian network were monitored, and the point at which the network suffered disconnection was detected, and the area where the network anomaly occurred was found. The correlation between the perceived results of internet structure change and the evaluation of real?world military conflict was analyzed, which provides information support for the observation of complex battlefield situation from internet space.