Feature Engineering and Machine Learning Framework for DDoS Attack Detection in the Standardized Internet of Things

NOTE : WORKS ONLY IN LINUX (UBUNTU)

Installation Guide

Drive link

Full Project avalable on ns-ddos

Project Overview

This project implements a feature engineering and machine learning framework for detecting Distributed Denial of Service (DDoS) attacks in the Internet of Things (IoT) environment. The framework utilizes sFlow, Floodlight, and Mininet for real-time detection.

Table of Contents

• Features
• Technologies Used
• System Design
• File Structure
• Installation
• Usage
• Dataset
• Real-Time Detection
• Results

Features

• Real-time DDoS detection using machine learning algorithms.
• Traffic sampling with sFlow.
• Network emulation with Mininet.
• Centralized control with Floodlight SDN controller.
• Feature extraction from network traffic data.

Technologies Used

• Python 3.6+
• Mininet
• Floodlight SDN Controller
• sFlow-RT
• Scikit-learn
• Pandas
• NumPy

System Design

System Design

File Structure

File Structure

Installation

1. Download File : ns-ddos

2. Set up Mininet: Follow the instructions on the Mininet website to install Mininet.

3. Set up Floodlight: Follow the instructions in the Floodlight configuration file (Floodlight Installation Steps) to configure Flood Light.

4. Set up sFlow-RT: Follow the instructions on the sFlow-RT website to install and configure sFlow-RT.

Usage

Follow the instructions in the Command.txt

Dataset

The dataset used for training and testing the machine learning models consists of network traffic data generated in the Mininet environment. The traffic data includes normal traffic as well as DDoS attack traffic.

Real-Time Detection

The ns-ddos file utilizes the trained machine learning model to detect DDoS attacks in real-time. It processes the incoming network traffic data and predicts whether it is normal or attack traffic.

Results

Project Snapshots

Dashboard	DDoS Protect
Metric Browser	Data Flow Test
Flow Trend	DDoS Protect Settings