Enhancing IoT Security RPL Attack Detection Using Sine Cosine Algorithm With XGBoost

Abstract

The most widely used routing protocol for Internet of Things (IoT) networks with limited resources is the (RPL). Network security is a significant issue because of the exponential growth of Internet of Things (IoT) devices and their growing ubiquity in safety-critical settings like healthcare and industry. One possible remedy for threat identification in these networks is using intrusion detection systems (IDS) based on machine learning. To achieve this, a machine learning approach that uses Random Forest (RF), k-nearest Neighbor (KNN), Decision Tree (DT), XGBoost, and Support Vector Machine (SVM) models is presented. The proposed machine learning-based detection approach conducts mitigate feature and classification using the Sine Cosine Algorithm (SCA) with XGBoost to select the smallest number of relevant features, leading to the best solution with the highest accuracy. The proposed model achieves a high accuracy of 97% on the Decrease Rank and Version Number. Also, a high accuracy of 99% on the Hello Flooding.