Simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) is a novel technology which enables the full-space coverage. In this letter, a multi STAR-RIS-aided system using non-orthogonal multiple access in an uplink transmission is considered, where the multi-order reflections among multiple STAR-RISs assist the transmission from the single-antenna users to the multi-antenna base station. Specifically, the total sum rate maximization problem is solved by jointly optimizing the active beamforming, power allocation, transmission and reflection beamforming at the STAR-RIS, and user-STAR-RIS assignment. To solve the non-convex optimization problem, a novel deep reinforcement learning algorithm is proposed which integrates meta-learning and deep deterministic policy gradient (DDPG), denoted by Meta-DDPG. Numerical results demonstrate that our proposed Meta-DDPG algorithm outperforms the conventional DDPG algorithm with $53\%$ improvement, while multi-order reflections among multi STAR-RISs yields to $14.1\%$ enhancement in the total data rate.

翻译：同时透射反射可重构智能表面（STAR-RIS）是一项实现全空间覆盖的新兴技术。本文研究了一种在上行传输中采用非正交多址接入的多STAR-RIS辅助系统，其中多个STAR-RIS之间的多阶反射协助单天线用户向多天线基站的传输。具体而言，通过联合优化主动波束成形、功率分配、STAR-RIS的透射与反射波束成形以及用户-STAR-RIS关联方案，解决了总速率最大化问题。针对该非凸优化问题，本文提出了一种融合元学习与深度确定性策略梯度（DDPG）的新型深度强化学习算法，称为Meta-DDPG。数值结果表明，所提出的Meta-DDPG算法较传统DDPG算法性能提升53%，而多STAR-RIS间的多阶反射能使总数据率提高14.1%。