Meta-Learning-Based STAR-RIS for Dynamic Multi-Mobile-User Downlink Communications Over 6G Mobile Wireless Networks

Reconfigurable intelligent surface (RIS) has been widely envisioned as a key technique, which can enhance communication quality of service (QoS) for mobile users (MUs) by reconfiguring wireless propagation environments. Unlike traditional RISs that can only reflect signals, simultaneously transmitting and reflecting RIS (STAR-RIS) can extend half-space coverage to full-space coverage by simultaneously transmitting and reflecting incident signals. In this paper, we propose the joint optimization for the transmission and reflection coefficients, i.e., phase-shifts and amplitudes, of STAR-RIS and the transmit beamforming of base station (BS) over STAR-RIS aided downlink communications, where MUs move in real time and can change their movement patterns dynamically. First, taking into account MUs' mobility, we formulate a rate maximization problem to maximize the average transmission rate between BS and multiple MUs. Then, since the traditional deep reinforcement learning (DRL) methods struggle to adapt to dynamic network environments, we develop a meta-learning based scheme to optimize STAR-RIS's transmission and reflection coefficients and BS's transmit beamforming. Finally, we validate and evaluate our developed scheme through extensive simulations, showing that our meta-learning based scheme significantly outperforms baseline schemes and can adapt to MUs' new movement patterns quickly.