Conference Name: Korean Institute of Communications and Information Sciences (KICS Winter 2024) Conference
Abstract: This study proposes an enhanced security scheme for industrial CPS by utilizing integrated blockchain, hierarchical federated learning (HFL), and aggregated signature techniques to develop a trusted and verifiable collaborative threat detection (CTD) framework. A verifiable off-on-chain aggregation mechanism is implemented to provide secure and tamper-resistant model aggregation with minimum transaction time, meeting the requirements of industrial services. Additionally, an efficient signature method is employed during the aggregation process to ensure the confidentiality and integrity of the model exchange.