Deep-Learning Based Reconfigurable Intelligent Surfaces for Intervehicular Communication

This paper proposes a novel deep neural network (DNN) assisted cooperative reconfigurable intelligent surface (RIS) scheme and a DNN-based symbol detection model for intervehicular communication. In the considered realistic channel model, the channel links between moving nodes are modeled as cascaded Nakagami-<inline-formula><tex-math notation="LaTeX">$m$</tex-math></inline-formula> channels, and the links involving any stationary node are modeled as Nakagami-<inline-formula><tex-math notation="LaTeX">$m$</tex-math></inline-formula> fading channels, where all nodes between source and destination are realized with RIS-based relays. The performances of the proposed models are evaluated and compared against the conventional methods in terms of bit error rate (BER) and computational complexity. It is shown that the proposed DNN-based systems achieve almost the same performance as conventional systems with low system complexity.