Two-Factor Authentication for IIoT Systems Utilizing Environmental Background Features

User authentication is critical for the secure operation of Industrial Internet of Things (IIoT) systems. This paper exploits background noise and images as an authentication factor during user login processes to enhance and improve the security of traditional authentication mechanisms. We first collect ambient noise and images of the user’s operational environment during the user’s routine work processes and extract the corresponding noise and image features in IIoT systems. Subsequently, we merge the noise and image features to create a novel feature space representing the user’s distinctive location factor. We further design ensemble classifiers based on three base classifiers KNN, SVM, and XGBoost for the new feature space, and determine the legitimacy of a user seeking access to the IIoT system. Finally, extensive experiments are conducted to illustrate the authentication performance and the related efficiency issues of the proposed authentication framework. The proposed work generates a dynamic authentication context, posing significant challenges for attackers. This approach not only enhances security but also resonates with the operational demands of industrial environments -being environmentally sensitive and requiring no additional investment.