Paper Title
Advancing Iot Security: Detecting Rpl Protocol Attacks With Multilayer Perceptron Networks

Abstract
As the Internet of Things (IoT) continues to expand across various sectors, it brings significant security challenges, especially in the context of the IPv6 Routing Protocol for Low-Power and Lossy Networks (RPL). This study presents a robust cybersecurity solution to counter internal routing attacks like Version Number (VN), Decreased Rank (DR), and DIS flooding attacks that exploit vulnerabilities within the RPL protocol. We propose an innovative detection framework utilizing a Multilayer Perceptron (MLP) model, which is fine-tuned with Sequential Feature Selection (SFS) to achieve high predictive accuracy while accommodating the resource limitations of IoT devices. The efficacy of our model is evaluated using the RPL-ELIDS dataset, where it exhibits exceptional accuracy rates: 98.06% for VN attacks, 99.33% for DR attacks, and 99.71% for DIS flooding attacks. These results highlight our model's enhanced ability to distinguish between benign and malicious traffic, surpassing the accuracy of existing models. Keyword - Internet of Things, RPL protocol, internal routing attacks, Multilayer Perceptron.