Description

March 11, 2024 · View on GitHub

This repository contains the tools used in our paper, Watching the Watchers: Practical Video Identification Attack in LTE Networks, accepted at USENIX Security 2022.

For more details, please check our paper

Dataset

Table 1: Dataset summary

Dataset# of Videos# of TracesDescription
YouTube10010029,715YouTube Top 100 in [1]
Netflix221,001Netflix Top 50
Amazon321,210Prime Video 32
YouTubeCA1007,383YouTube Top 100 in [1] (w/ CA)
YouTube2002006,424YouTube Top 101-300 in [1]
Web-268Visiting randomly selected 45 websites from Alexa Top 50 Website [2]
Teleconf-201Google Meet

[1]: Most viewed music videos of all time. https://kworb.net/youtube/topvideos.html.

[2]: Katharina Kohls, David Rupprecht, Thorsten Holz, and Christina P̈opper. Lost Traffic Encryption: Fingerprinting LTE/4G Traffic on Layer Two. ACM WiSec19

Table 2: # of Traces used in the evaluation (YouTube100 videos)

MNOVideo Quality# of TracesMNOVideo Quality# of TracesMNOVideo Quality# of Traces
A480p3,184B480p3,376C480p3,262
A720p3,196B720p3,318C720p3,218
A1080p3,645B1080p3,223C1080p3,293

How to use

Environment Setup

You need to prepare at least one labtop, one cellphone, and one Linux server.

Labtop Setup

Set up Innowireless DM tool on a labtop and connect a cellphone used to collect data. Most of the Python scripts will be run on this labtop.

Linux Server Setup

Set up airscope on a Linux server. We use airscope to capture radio signals in the same way as an adversary.

Additionally, please copy send_mail.py to the Linux Server and install smtpd on the server. This script will be run to notify you when experiments are completed via email.

Configuration

Many configuration variables depend on your experimental environment. Please see config.py and modify the variables appropriately.

Authors

This project has been conducted by the below authors at KAIST.

Citation

We would appreciate if you consider citing our paper when using the tools.

@inproceedings{bae:2022:watching,
  author = {Sangwook Bae, Mincheol Son, Dongkwan Kim, CheolJun Park, Jiho Lee, Sooel Son, and Yongdae Kim},
  title = {Watching the Watchers: Practical Video Identification Attack in LTE Networks},
  booktitle = {Proceedings of the 31st USENIX Security Symposium (Security)},
  mon = aug,
  year = 2022,
  address = {Boston, MA}
}