Optimization Accuracy of Blackhole Attacks in RPL Network using Machine Learning Technique

Abstract

Abstract: - Routing Protocol for Low-Power and Lossy Networks (RPL) is widely adopted for Internet of Things (IoT) applications due to its energy efficiency and scalability. However, RPL is vulnerable to various security threats, particularly blackhole attacks, where a malicious node drops all the packets it receives, leading to significant performance degradation. This research focuses on optimizing the detection and mitigation of blackhole attacks in RPL networks using machine learning techniques.

The study leverages supervised machine learning algorithms to classify and identify malicious nodes based on network traffic patterns, packet loss rates, and routing metrics. Feature selection methods are employed to enhance model performance by identifying the most relevant parameters influencing blackhole detection. Comprehensive experiments are conducted using a simulated RPL network environment to evaluate the proposed approach in terms of accuracy, precision, recall, and F1-score.