一种基于云平台的数据流量监控方法及系统
By deploying traffic probes at the cloud platform entry point for in-depth analysis and time series analysis, combined with an improved isolated forest algorithm and classification model, the threshold is dynamically adjusted to identify abnormal traffic and issue defense rules. This solves the problems of lack of metadata and inaccurate protection in cloud platform network traffic monitoring, and realizes an automated anomaly detection and protection closed loop.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 陕西安康玮创达信息技术有限公司
- Filing Date
- 2026-05-22
- Publication Date
- 2026-07-17
AI Technical Summary
The existing cloud platform network traffic monitoring architecture lacks deep analysis capabilities and cannot fully extract metadata. Traditional anomaly detection algorithms cannot adapt to traffic fluctuations, and the protection system lacks an automated processing mechanism, resulting in inaccurate identification and protection of abnormal traffic and an inability to form a complete closed loop.
Traffic probes are deployed at the cloud platform entry point to identify protocols and perform deep analysis, generate metadata sets, and generate traffic time series by aggregating them through time windows. An improved isolated forest algorithm is used to dynamically adjust thresholds, and a pre-trained classification model is combined to identify abnormal traffic types. Defense rules are then dynamically distributed to traffic cleaning devices for filtering and processing.
It enables accurate anomaly identification and type determination of network traffic on cloud platforms, automates the matching and execution of protection strategies, and ensures the standardized operation of network transmission.
Smart Images
Figure CN122247769B_ABST