/stream-monitoring-app

Analyzing Encrypted Network Traffic via TLS Interception

Primary LanguagePython

stream-monitoring-app

It's my senior project displaying watched video content from Youtube and Udemy by the LAN devices by performing arp poisoning attack.

Attention

Last commit containing bar charts and graphical representations about the total downloads by specific device is missing.

Project Description

In this project, we analyzed the encrypted network traffic between video streaming platforms which are Udemy and Youtube, and clients by using an HTTPS proxy. We enhanced a system handling each request to the video platforms by target devices in the local area network. We forwarded network traffic of each target device to the admin computer by performing ARP poisoning attack. Then we analyzed all network flow of multiple devices by running multiple scripts that are responsible for identifying network packets and then send them to the MongoDB database instance. Afterwards, we implemented a web application displaying each client’s video history in terms of viewed video title, bandwidth usage during watching the video, statistical graphs such as total download amount in this week for each device. The system works like a network packet sniffer but we are filtering network packets by looking for urls that are requested by clients and only we are capturing, parsing and storing Youtube and Udemy network traffic.

Project Report and Demo Links

Technologies used

arpspoof

HTTPS Proxy

  • mitmproxy Python API

Frontend

  • React.js
  • Bootstrap
  • Sass
  • Yarn

Backend

  • Express.js

Database

  • MongoDB Atlas