Proposing a Rank and Wormhole Attack Detection Framework using Machine Learning

Internet of Things (IoT) is a paradigm in digital technology which is prevalently revolutionizing various sectors like healthcare, military, business and more. However, the incremental deployment of this advanced technology has also caused critical security issues simultaneously. In particular, IoT networks are continuing to grow vulnerable to security attacks due to exponential connectivity of ‘things’ with each other in the smart infrastructure. Due to this increased vulnerability, it has become crucial to address the issue of insecure routing in these IoT devices. IoT uses RPL, which is a specially designed standard for networking that caters to the resource-constrained and lightweight nature of IoT devices, for information broadcast. It is equally prone to routing attacks like any other class of protocols in wireless networks. Various solutions have been proposed by researchers to counter them including version, rank, sinkhole and wormhole attacks since last decade. However, given the huge impact, neither detection nor mitigation method has been found which addresses rank and wormhole attacks when they are initiated at the same time on an IoT network. In this paper, a rank and wormhole attack detection framework is proposed, by employing machine learning approaches, which address the stated issue. This research aims to contribute toward design and development of high-performing and effective solutions for routing attacks in RPL-based IoT networks.