https://github.com/erikglz/coap-mtd

Repository for an IoT security project implementing Moving Target Defense (MTD) through CoAP protocol randomization to mitigate spoofing attacks and enhance adaptive security.
https://github.com/erikglz/coap-mtd

coap-protocol cybersecurity iot machine-learning python scikit-learn spoofing

Last synced: 4 months ago
JSON representation

Repository for an IoT security project implementing Moving Target Defense (MTD) through CoAP protocol randomization to mitigate spoofing attacks and enhance adaptive security.

• Host: GitHub
• URL: https://github.com/erikglz/coap-mtd
• Owner: ErikGlz
• License: mit
• Created: 2025-02-24T01:39:45.000Z (4 months ago)
• Default Branch: main
• Last Pushed: 2025-02-24T16:22:48.000Z (4 months ago)
• Last Synced: 2025-02-24T17:35:30.742Z (4 months ago)
• Topics: coap-protocol, cybersecurity, iot, machine-learning, python, scikit-learn, spoofing
• Language: Python
• Homepage:
• Size: 429 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# CoAP-MTD

## Overview  

This project explores security in **Internet of Things (IoT)** systems by implementing a **Moving Target Defense (MTD)** technique. The approach involves continuously reconfiguring the **CoAP (Constrained Application Protocol)** communication to mitigate **spoofing attacks** and other network threats.  

By randomizing CoAP protocol dialects and integrating machine learning-based decision-making, the system enhances resilience and adaptability against cyber threats in IoT environments.  

## Features  

- **CoAP Protocol Dialect Randomization** – Implemented using the `aiocoap` Python library.  

- **Dynamic Reconfiguration** – The system continuously modifies CoAP communications to increase security.  

- **Machine Learning Integration** – Uses a **decision tree algorithm** from `scikit-learn` to analyze and adapt security measures.  

- **IoT System Implementation** – Built with a **Raspberry Pi 4** as a sensor node and a **Linux client** for communication.  

- **Attack Simulation** – Uses **Scapy** to generate and test spoofing and network-based attacks.  

## Technologies Used  

- **Python** (Core language)  

- **aiocoap** (CoAP protocol implementation)  

- **scikit-learn** (Decision tree algorithm)  

- **Scapy** (Network attack simulation)  

- **Raspberry Pi 4** (IoT sensor node)  

- **Linux** (Client system)

## Future Work

- Enhancing the adaptability of the MTD technique.

- Expanding machine learning models for better threat detection.

- Testing with additional IoT devices and protocols.