Improved Intrusion Detection in the Internet of Things: A Multi-Layered Neural Network Approach and Analysis
Mansoor Farooq, Faheem Ahmad · International Journal of Computing · 2024
The (IoT) Internet of Things is a complex notion that refers to the interconnection of several individual devices over a network (IoT). The data gathered by these interconnected devices have the potential to have far-reaching consequences for human society, the economy, and the environment. The IoT is especially vulnerable to a variety of vulnerabilities in hostile environments like the internet. High-end security solutions are not adequate to safeguard an IoT system due to adequate storage and less processing capabilities. This emphasizes the need for ascendable, strewn, and robust smart security solutions. In this study, IoT networks are safeguarded depleting a multiple-layered security strategy centered on deep learning. The proposed architecture employs the use of three intrusion detection datasets CIC-IDS, BoT-IoT, and ToN-IoT to weigh the performance of the insinuated multiple-layered approach. Irrevocably, compared to 92% accuracy for the existing methodologies, the new layout obtained 98% accuracy.