PyTorch Network Analyzer

Getting Started

1. Prerequisites

• Ensure you have Git installed. Download it from git-scm.com.

2. Cloning the Project

Clone the repository using Git Bash or command line:

git clone https://github.com/nmsu-senior-project/pytorch-network-analyzer.git

3. Setting Up Python Environment

Navigate to the project directory and create a Python virtual environment:

cd pytorch-network-analyzer
python -m venv venv

Activate the virtual environment:

• Windows:

  source venv/Scripts/activate

• Linux:

  source venv/bin/activate

4. Installing Dependencies

Install required Python libraries:

pip install scapy mysql-connector-python

5. Setting Up MySQL Server

• Download and install MySQL from dev.mysql.com.
• Create a local MySQL server. Use tools like DBeaver for GUI management or command line for manual setup.

MySQL Command-Line Interface (CLI) Setup on Windows

Adding MySQL to PATH Environment Variable

1. Find your MySQL Installation Directory (e.g., C:\Program Files\MySQL\MySQL Server 8.0\bin).
2. Add MySQL to PATH:
   • Press Win + X and select System.
   • Click on Advanced system settings -> Environment Variables.
   • Under System variables, edit Path and add your MySQL bin directory.
   • Restart Command Prompt to apply changes.

Verifying MySQL Installation

Open a new Command Prompt window and verify MySQL is recognized:

mysql -u root -p

MySQL Database and User Creation Commands

Setting Up Database Credentials

Create a credentials.txt file in the analyzer directory with MySQL user credentials:

db_user:user1
db_pass:password1
db_user:user2
db_pass:password2
db_user:user3
db_pass:password3

Project Overview

Purpose

The project aims to develop a PyTorch-based deep learning model to analyze network traffic and detect common threat behaviors, enhancing cybersecurity measures.

Workflow

1. Data Processing and Baseline Creation:

   • Process packet traffic from PCAP files.
   • Store baselines in MySQL for device network activity.

2. Threat Analysis Using PyTorch:

   • Train a model to monitor and alert on unusual activities.

Future Goals

• Automate batch processing of PCAP files.
• Enhance PCAP file management for deeper analysis.
• Develop a Django web interface for model visualization.

Summary

This project combines baseline data with PyTorch's capabilities to provide robust network security management tools, detecting and addressing potential threats effectively.