ML-Driven Attack Detection in RPL Networks: Exploring Attacker Position's Significance

The Routing Protocol for Low Power and Lossy Networks (RPL) plays a pivotal role in IoT communication, employing a rank-based topology to guide routing decisions. However, RPL is vulnerable to Decreased Rank Attacks, where malicious nodes illegitimately lower their ranks to manipulate routing paths. While exploring the applicability of machine learning (ML) techniques for attack detection holds promise, their effectiveness is often overlooked in the context of attacker position within the network. This study bridges this gap and delve into investigating the impact of attacker position on Decreased Rank attack detection using ML-based approaches. Our findings reveal that the success of attack detection is highly contingent on the attacker's proximity to the network root, highlighting the importance of considering network topology in attack mitigation strategies.