A Novel Deep Learning-based Framework for Blackhole Attack Detection in Unsecured RPL Networks

The routing protocol for low-power and lossy networks (RPL) was developed specifically for constrained communication. Considering its constrained nature, RPL-based Networks can be accessible by trusted and untrusted global users via the Internet and can be subject to serious attacks. Routing attacks are especially difficult to be identified when they occur. However, Deep Learning techniques can be leveraged in detecting network intrusions. This paper comes up with a new deep learning-based framework for routing attack detection in unsecured RPL networks. It allows analyzing and processing the network traffic, extracting features, and defining target-based intrusion thresholds, which leads to the detection of routing attacks. The proposed model is compared to the baseline Machine learning methods. Extensive simulation results confirm the efficiency of our proposed model with a reliable error rate and a detection accuracy up to 98.70%.