/pcap-data-processing

Scripts used to process and evaluate packet captures, and dataset generation scripts for machine learning tasks

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

PCAP Data Processing

Scripts used to process and evaluate packet captures.

Installation

There is a convenience install script in the root directory of the repo.

This can be run with the following command: ./install.sh

Tests

  1. pkt2flow:
cd flow-splitting/pkt2flow
./pkt2flow -h
  1. nDPI:
cd labelling/nDPI/tests
./do.sh

Running the Code

Ensure you have run the installation script and installed the required python packages found in requirements.txt. Please read the README files in each directory if you are running into errors and ensure you set your environment variables in each folder.

Creating and Labelling Flows

  1. Navigate to the flow-splitting directory and run ./split.sh.
  2. Navigate to the labelling directory and run ./label.sh.
  3. Format labels by running python3 clean_label.csv.py in the labelling directory.