RIS-UAV Assisted Marine Communication Based on Deep Neural Network Training

With the reconfigurable intelligent surface (RIS) and unmanned aerial vehicle (UAV) becoming increasingly prevalent in everyday human activities, RIS-UAV assisted wireless communication system helps to extend transmission distance and improve communication quality. However, most of the scenarios about RIS-UAV applications are on land. At the same time, how to jointly optimize the communication link between the UAV equipped with RIS and the communication link between the base station (BS) and the vessel is challenging. In order to solve the above problems, this paper proposes a RIS-UAV assisted marine communication system, and training the beamforming matrix, transmitter and receiver weights implemented by deep neural networks (DNN) to maximize the performance of the system. The simulation results show that the marine wireless communication system based on RIS-UAV achieves lower bit error rate (BER) than the conventional UAV relay communication system. Furthermore, based on parallel intelligence theory, more reasonable intelligent decision-making and prediction are expected be obtained for marine communications in the future.