Securing the Internet of Things using AI-Enabled Detection of Attacks via Port Scans in IoT Networks

As the Internet of Things (IoT) rapidly gains popularity, preserving the security and dependability of IoT networks has become crucial. Having a security system that is both robust and intelligent is crucial because port scan attacks, which are growing more prevalent and may take advantage of vulnerabilities in Internet of Things devices, can be used. This research paper demonstrates a novel approach to protecting the Internet of Things by using AI-powered threat detection techniques. These techniques primarily aim to thwart port scanning in IoT networks. The proposed solution that completely analyzes and groups network data using cutting-edge machine learning methods. This helps in determining which connections are secure and which might be risky. The system can instantly identify any issues with port scan operations since it monitors the data and extracts the crucial portions. The proposed solution for detecting attacks through port scans in Internet of Things networks, businesses are able to improve the security of their IoT infrastructure. Aposemat 1oT-23 provided the data set, which was utilized to look into a malicious attack that was launched on an IoT device using Port Scan. The CNN model, the Naive Bayes model, and the ML model for logistic regression are all used in this work. This research demonstrates how effectively port scans combined with the CNN approach may detect impending attacks on IoT devices early on. The results demonstrate that, in addition to the F1 score, performance in terms of precision, recall, and accuracy has greatly increased with CNN.