Intelligent Event-Triggered Privacy-Preserving Consensus of Connected Vehicle Group Systems in Open RANs Under Stochastic FDI Attacks

The open radio access network (RAN) technology is crucial to the realization of autonomous driving, however, it also leads to the emergence of privacy protection and network security issues. This article explores the privacy-preserving consensus tracking problem of connected vehicular group systems over open RANs subject to stochastic false data injection (FDI) attacks. First, based on the polytope linearization technique and graph theory, a distributed connected vehicle group system model with multiple uncertain parameters is established under the two-layer cyber-physical framework. Then, by resorting to the homomorphic encryption method, a novel privacy-preserving consensus control strategy is designed via a vanishing affine output mask function, which not only prevents the leakage of initial state information but also improves the resilience of vehicle state to FDI attacks. Furthermore, an intelligent event-triggered network transmission scheme is proposed by introducing the Q-learning algorithm to dynamically adjust the triggering threshold during each iteration, which greatly lightens the communication load of open RANs. Finally, an illustrative example is presented to verify the validity of the derived theoretical results.