FADIC: a fine-grained attribute-based dynamic group key agreement scheme for IoT

Abstract The Internet of Things (IoT), serving as a principal technology enabling device interconnection, data amalgamation, and automation, boasts unrivaled ubiquity and convenience. Due to limited computing resources, small storage space, and restricted security features of IoT terminals, they are exposed to security risks such as information leakage, tampering, unauthorized access, and malicious control. To assure the security of data communication within the IoT framework necessitates the deployment of a Group Key Agreement (GKA), and the categorization of devices into specific groups. However, due to the dynamic structure of the IoT, which demands frequent device node addition and removal, might engender the constant need for group key updates. Traditional GKA lacks sufficient flexibility and response speed, resulting in additional computing and communication costs. Subsequently, group key updates could also pose a challenge to the forward and backward security provided by the GKA. In this work, we take a feasible step towards this challenge by proposing FADIC, a fine-grained attribute-based dynamic group key agreement scheme for IoT. We employ a server-assisted approach for the generation and administration of access trees, and achieve an efficient attribute-based dynamic group key agreement scheme devised by integrating Lagrange interpolation and attribute-based access tree structure. Our FADIC scheme ensures fine-grained access control and privacy protection for group members. Moreover, the server-assisted approach is utilized to compute the updated keys, resulting in a substantial reduction in communication and computational overheads associated with attribute-based group key agreement schemes. We formally prove that our FADIC is indistinguishable under chosen ciphertext attacks secure, supporting the desired forward and backward security. The comparison results unequivocally demonstrate that our FADIC outperforms its foremost counterparts.