There emerges a strong trend recently to create an intelligent and controllable wireless transmission environment by installing reconfigurable intelligent surfaces (RIS) on the surface of diverse. We research the use of RIS in 5G multicast TV applications in this paper. In particular, the shape-adaptive RIS is developed to relay 5G multicast TV signals and achieve extra RIS gain. To optimize the shape bending in RIS model, we employ the algorithm of Deep Deterministic Policy Gradient (DDPG), a reinforcement learning technique that combines both Q-learning and Policy gradients. Our simulation results show that RIS can considerably enhance the SINR of worst users and improve the system’s overall Modulation and Coding Scheme (MCS) level for 5G mobile broadcasting services.

中文翻译：

---

利用形状自适应 RIS 增强 5G 移动广播服务


最近出现了一种强烈的趋势，即通过在各种物体表面安装可重构智能表面（RIS）来创建智能、可控的无线传输环境。我们在本文中研究了 RIS 在 5G 组播电视应用中的使用。特别是，形状自适应RIS被开发用于中继5G组播电视信号并实现额外的RIS增益。为了优化 RIS 模型中的形状弯曲，我们采用了深度确定性策略梯度（DDPG）算法，这是一种结合了 Q 学习和策略梯度的强化学习技术。我们的仿真结果表明，RIS 可以显着提高最差用户的 SINR，并提高 5G 移动广播服务的系统整体调制和编码方案 (MCS) 水平。