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.
- CoAP Protocol Dialect Randomization – Implemented using the
aiocoapPython library. - Dynamic Reconfiguration – The system continuously modifies CoAP communications to increase security.
- Machine Learning Integration – Uses a decision tree algorithm from
scikit-learnto 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.
- 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)
- Enhancing the adaptability of the MTD technique.
- Expanding machine learning models for better threat detection.
- Testing with additional IoT devices and protocols.