This project is part of my bachelor thesis in Computer Science.
ISSS is a cybersecurity tool designed for IoT devices. The tool utilizes a machine learning model to analyze generated network traffic. Upon detecting malicious traffic, the tool identifies the actors involved in the communication and adds a firewall rule for these devices to isolate them from the network.
The idea behind the project originated from a study on ransomware attacks, a form of extortionate cyberattack which nowadays represents one of the most significant threats to small and medium-sized enterprises, and it is predicted to increasingly involve ordinary individuals through the compromise of smart devices that are becoming more widespread within households.
The IoT-Subnetting Security System aims to secure the network of IoT devices within homes.
The tool operates by using a third-party device acting as an access point to which IoT devices can be associated. Once connected, the network flows generated by the various devices are evaluated by a machine learning model (currently using a random forest) to determine their legitimacy. If a flow is classified as malicious, the tool ensures that the actors involved in the communication are isolated, preventing them from continuing to interact and potentially spreading the infection or launching further attacks on third parties.
Currently, the tool is designed to be installed on a Raspberry Pi-3 module, but in the future, the idea is to make it configurable on different architectures as well.
The project can be found on my GitHub page (https://github.com/SebastianoMorson/IoT-Subnetting-Security-System-v2).
Here can be found my thesis paper. Is written in italian, but now Google Translate can translate entire documents, so if you want to read in other languages, feel free to translate it.