Routing Security in RPL-based Internet of Things (IoT) Networks using Whale Optimisation Algorithm and Unsupervised Machine Learning

Routing Protocol for Low-Power and Lossy Networks is a widely adopted routing protocol specifically designed for Low Power and Lossy Networks. It offers energy-efficient mechanisms and supports secure modes of operation, making it suitable for resource-constrained devices in IoT. So, RPL is extensively utilized in various IoT applications for efficient data communication. However, the issue of routing security in RPL-based IoT networks has emerged as a significant concern. To address this, Whale Optimization Algorithm is used to select the relevant features to reduce the resource consumption. The selected relevant features are given as input to various unsupervised Machine Learning models. Among the models, Local outlier factor yielded the inlier accuracy of $\mathbf{9 9. 3 0 \%}$ and outlier accuracy of $\mathbf{9 9. 6 8 \%}$. The results are also compared with and without feature selection algorithm.