Adopting Machine Learning Blockchain Intrusion Detection for Protecting Attacks on Internet of Things

Abstract

Intrusion Detection Systems (IDSs) are widely used in various computer networks with the goal of spotting cyber threats and potential incidents. Collaborative intrusion detection networks (CIDSs) have been developed to augment the detection power of a single IDS by allowing IDS nodes to exchange data. The Internet of Things (IoT) can be thought of as a network or connectivity of sensors and actuators that share data in a unique way. Blockchain technology has been applied in a variety of fields to foster trust and data protection by enabling participants to trade transactions and communicate information while preserving a level of trust, integrity, and greater transparency. However, there are numerous security concerns associated with the implementation architectures and technologies that will form the Internet of Things' backbone. Hence, this paper proposes a machine learning technique leveraging on blockchain technology with IDS for detecting attacks on IoT. In this paper, we used Naïve Bayes (NB), K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) models for performing the experiment on NSLKDD dataset. The experimental findings for KNN model achieved 99.6% detection rate with a false alarm rate of 0.4. The NB and SVM models also gave competitive results. Keywords: Machine Learning, Blockchain, Intrusion Detection System, Internet of Things, K-Nearest, Online Safety, Neighbor, Collaborative IDS