Blazer: Encrypted Video Traffic Identification for Mixed Segment Transmission Pattern based on LLM

作者：Weitao Tang, Meijie Du, Die Hu, Li Shu, Z.C. Li, Rong Yang, Qingyun Liu · 年份：2026 · DOI：10.1145/3805622.3810657 · 研究领域：Internet Traffic Analysis and Secure E-voting、Digital Media Forensic Detection、Video Analysis and Summarization

Determining the source of encrypted video traffic is an important task in network regulation. In the context of Dynamic Adaptive Streaming over HTTP (DASH), the newly emerged mixed segment transmission pattern introduces substantial difficulties for fingerprint matching, especially under adverse network conditions. To address these challenges, we propose Blazer, a DASH encrypted video traffic identification method for the mixed segment transmission pattern. First, we design a novel fingerprint that integrates video and audio segment sequences. Then, we extract the traffic fingerprint from the TLS record layer of video traffic. Finally, by observing implicit segment-mixing constraints, we design a targeted prompt and Retrieval Augmented Generation (RAG) that enables Large Language Models (LLMs) to perform fingerprint matching effectively. Across 12 network scenarios, Blazer delivers substantially better performance than the other 4 SOTA methods.