Beam Training and Codebook Design for RIS Assisted UAV Communications in Emergency Rescue

Reconfigurable intelligent surfaces (RIS) assisted unmanned aerial vehicle (UAV) communications are an effective way to enhance communication and effectively improve rescue efficiency in disaster scenarios. This provides a good communication guarantee for the collection and transmission of big data. Beam training is the key method to solve the problem of beam alignment between the receiving and transmitting ends. However, existing schemes rely on feedback from uniform or finite precision codebooks, which are not suitable for complex electromagnetic environments, resulting in large beam training overhead. We consider a non-uniform codebook-based beam training scheme under Karush-Kuhn-Tucker (KKT) conditions to optimize the energy consumption of rescued user. Specifically, we consider the RIS assisted UAV communication system with emergency rescue. Then, considering the constraints of RIS phase-shift, the system communication rate and energy efficiency, we propose an optimization problem to minimize the system transmission power. In addition, we propose an optimization algorithm of successive approximation codebook iteration with Karush-Kuhn-Tucker (KKT) to solve this problem with low precision non-uniform codebook. Finally, the simulation results show that the proposed optimization algorithm can effectively reduce the transmission power of the system.