Secure and efficient authenticated group key agreement protocol for AI-based automation systems.

Automation systems are soaring due to the innovation of artificial intelligence (AI) technologies. In this paper, we mainly concentrate on the security and efficiency of data transmission in AI-based automation systems, especially, for data sharing in a group manner in distributed networks. To this end, an authenticated group key agreement protocol is proposed for secure data transmission in AI-based automation systems. In order to release the computational overhead of distributed nodes, a semi-trusted authority (STA) is introduced to enable precomputation operations. Moreover, to overcome the predominantly distributed denial of service (DDoS) attack, a dynamic batch verification mechanism is devised. The presented dynamic batch verification mechanism guarantees that the proposed protocol can be run properly among legitimate nodes regardless of the existence of nodes who suffered from the DDoS attack. Finally, the session key security of the proposed protocol is proved and the performance is evaluated.