在COVID-19场景中主动检测e-learning系统中网络异常的方法建议
Methodology proposal for proactive detection of network anomalies in e-learning system during the COVID-19 scenario
摘要：
在特定条件和危机情况下，如冠状病毒大流行（SARS-CoV-2）或COVID-19疾病，电子学习系统对于顺利执行教学和其他教育过程至关重要。在这种情况下，电子学习生态系统要素的可用性得到进一步强调。从对AAI@EduHr的DDoS（分布式拒绝服务）攻击中可以明显看出确保这种生态系统可用性的重要性，AAI@EduHr是克罗地亚共和国许多电子学习用户的关键认证服务。在这样做的过程中，许多用户（教师/学生/管理员）被阻止执行和参与计划的教学过程。鉴于DDoS作为一种异常的网络流量已被确定为在危机情况下对电子学习生态系统的主要威胁之一，本研究将侧重于概述的方法，去病毒化模型的主动检测DDoS流量。检测中的挑战是有效区分由合法用户活动（flash crowd）引起的增加的流量强度和服务请求与由DDoS攻击引起的非法流量。通过以下分析方法开发的DDoS流量检测模型将作为基础，以响应可能对电子学习生态系统元素（如DDoS攻击）的可用性产生负面影响的事件的形式提供进一步的指导方针和建议。
英文摘要：
In specific conditions and crisis situations such as the pandemic of coronavirus (SARS-CoV-2), or the COVID-19 disease, e-learning systems be-came crucial for the smooth performing of teaching and other educational pro-cesses. In such scenarios, the availability of e-learning ecosystem elements is further highlighted. An indicator of the importance for securing the availability of such an ecosystem is evident from the DDoS (Distributed Denial of Service) attack on AAI@EduHr as a key authentication service for number of e-learning users in Republic of Croatia. In doing so, numerous users (teach-ers/students/administrators) were prevented from implementing and participat-ing in the planned teaching process. Given that DDoS as an anomaly of network traffic has been identified as one of the key threats to the e-learning ecosystem in crisis scenarios, this research will focus on overview of methodology for de-veloping a model for proactive detection of DDoS traffic. The challenge in de-tection is to effectively differentiate the increased traffic intensity and service requests caused by legitimate user activity (flash crowd) from the illegitimate traffic caused by a DDoS attack. The DDoS traffic detection model developed by following analyzed methodology would serve as a basis for providing further guidelines and recommendations in the form of response to events that may negatively affect the availability of e-learning ecosystem elements such as DDoS attack.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
