Inferring YouTube Streaming QoE from Encyrpted Traffic
Video
Team Information
Team Members
Craig Gutterman, PhD Candidate, Department of Electrical Engineering, School of Engineering and Applied Science (SEAS), Columbia Engineering
Trey Gilliland, Undergraduate Student, Department of Computer Science, School of Engineering and Applied Science (SEAS), Columbia Engineering
Gil Zussman, Professor, Department of Electrical Engineering, School of Engineering and Applied Sciences (SEAS), Columbia Engineering
Ethan Katz-Bassett, Associate Professor, Department of Electrical Engineering, School of Engineering and Applied Science (SEAS), Columbia Engineering
Katherine Guo, Research Scientist, Nokia Bell Labs
Sarthak Aruro, Undergraduate Student, Department of Electrical Engineering, School of Engineering and Applied Science (SEAS), Columbia Engineering
Faculty Advisor: Gil Zussman, Professor, Department of Electrical Engineering, School of Engineering and Applied Sciences (SEAS), Columbia Engineering
Abstract
As video traffic for services such as YouTube and Netflix dominate the Internet, it is important for operators to monitor video playback Quality of Experience (QoE) metrics to provide users with the best possible viewing experience. However, with wide deployment of encrypting video traffic information, this task has become near impossible for internet service providers. This poses a challenge for network operators to monitor user QoE and improve upon their experience. To resolve this issue, we develop and present a system for REal-time QUality of experience metric detection for Encrypted Traffic, Requet. Requet uses a Machine Learning model to identify video and audio chunks from the IP headers of encrypted traffic which is then used to predict quality of experience metrics such as video quality, state, and buffer health. With this information, network providers can monitor video playback more effectively and provide a superior streaming experience. Contact this Team
Team Contact: Trey Gilliland (use form to send email)
