An autonomous intrusion detection system for the RPL protocol

Routing Protocol for Low-Power and Lossy Networks (RPL) is a proactive routing protocol for wireless networks based on distance vectors operating on the platform of IEEE 802.15 (a working group of the Institute of Electrical and Electronics Engineers (IEEE) IEEE 802 standards committee). Various approaches have been proposed to detect intrusions in the RPL, often accompanied by problems such as inaccuracy and high error in intrusion detection. Therefore, in this paper, a combination of the hierarchical semantic approach and the Group Method of Data Handling (GMDH) neural network algorithm was used to detect intrusions. The hierarchical semantic approach, which is based on the conversion of infiltration values into meaningful numbers, and the selection of influential variables in the infiltration of the Internet of Things can significantly affect intrusion detection. The GMDH algorithm also produces a model based on the neural network with hidden layers. It also learns from past events and identifies possible intrusions that may occur in the future. The results obtained in the proposed model were compared with the results of other methods. It was observed that the accuracy of the proposed method improved by about 20%.