MininetIDS

MininetIDS is a consolidated environment for training, testing, and deploying Machine Learning-based Intrusion Detection Systems (ML-IDS) in Software-Defined Networks (SDN). It leverages Mininet for network emulation and the Ryu controller for SDN management, providing a comprehensive platform for researchers and network security professionals to develop and evaluate advanced IDS solutions in SDN environments.

Features

• Dataset management (import, select, analyze, preprocess)
• Machine learning model training (Logistic Regression, KNN, Naive Bayes, Decision Tree, Random Forest)
• Network topology management
• Integration with Ryu controller for IDS functionality
• Feature selection and data preprocessing tools

Included Datasets

MininetIDS comes with two pre-included datasets for testing and evaluation:

1. MininetFlows

   • Generated using Mininet
   • Contains DoS attacks:
     • ICMP flood
     • TCP SYN flood
     • UDP flood
     • LAND attack
   • Includes normal traffic:
     • HTTP
     • TCP
     • UDP
     • ICMP

2. NSL-KDD

   • Widely used benchmark dataset for network intrusion detection research
   • Improved version of the original KDD Cup 1999 dataset
   • Contains various types of network attacks and normal traffic

These datasets provide a starting point for testing and evaluating IDS models within the MininetIDS environment. The MininetFlows dataset offers simulated traffic that closely matches the Mininet environment, while NSL-KDD provides a standard benchmark for comparison with other IDS research.

Installation

1. Update your system:

   sudo apt-get update
   

2. Install Git:

   sudo apt-get install git
   

3. Clone the repository:

   git clone https://github.com/ranauzairahmed/MininetIDS.git
   

4. Navigate to the project directory:

   cd MininetIDS
   

5. If you already have Mininet and Ryu working properly:

   pip install -r requirements.txt
   

6. Otherwise, make the installation script executable:

   chmod +x install.sh
   

7. Run the installation script:

   ./install.sh
   

Usage

To start the MininetIDS interface, run:

python3 MininetIDS.py

This will launch the command-line interface where you can use various commands to manage datasets, train models, and control the IDS.

For a full list of commands, use the help command in the MininetIDS interface.

Example Demonstration

NSL-KDD with MininetIDS

License

MIT License

Associated with Capital University of Science & Technology

Faculty of Computing
BSCS - Final Year Project

Supervisor:

Dr. Muhammad Siraj Rathore
[email protected]
Faculty Profile
Google Scholar
ResearchGate

Group:

Rana Uzair Ahmed
[email protected]
LinkedIn

Raja Tayyab Nawaz
[email protected]
LinkedIn

Reference

SDN Network DDoS Detection Using Machine Learning