Anomaly Detection Algorithm with Blockchain to Detect Potential Security Attacks in the IIoT Model of Industry 5.0

The research presents a model for detecting potential security attacks in the Industry 5.0’s Internet of Things (IIoT) model using an Anomaly Detection Algorithm, with Blockchain technology to further enhance security. One-class Support Vector Machines (SVM) is used as the Anomaly Detection Algorithm, to identify any unusual behavior in the IIoT system. The proposed model ensures the integrity of data by implementing the decentralized features of Blockchain technology. This paper aims to address the current security challenges faced by Industry 5.0 and enhance the reliability of the IIoT model. Since Industry 5.0 is not here yet, hypothetical data is used to train the model which is generated after seeding using Numpy. The Blockchain technology enhanced the overall security of the Industrial Internet of Things (IIoT) model whereas, to secure it even further by detecting anomalous activities, the machine learning algorithm is proposed. Anomaly detection algorithm with Gaussian distribution is proposed through One-class SVM. The threshold for an activity to be classified as unusual or anomalous is discussed in the paper along with the difference between classification algorithm and anomaly detection algorithm. The research implemented One-class SVM algorithm to train the model by randomly seeding data using Numpy with an average accuracy of 92.8% after 5 different runs with different datasets. The algorithm also focused on other applications of the model like detection of faulty driver, device, or equipment.