Resource Allocation in STAR-RIS-Aided SWIPT with RSMA via Meta-Learning

TL;DR

This work tackles the joint optimization of STAR-RIS aided SWIPT in a downlink setting where users are split into information decoding receivers and energy harvesting receivers and served via RSMA. The authors formulate a non-convex objective balancing the normalized sum-rate and harvested energy, and solve it with a Meta-DDPG approach that adapts beamforming, STAR-RIS phase shifts, and common-rate allocation in dynamic wireless environments. The Meta-DDPG framework combines policy learning with meta-learning to enable rapid adaptation, using a reward structure based on $R_{ ext{Norm}}$ and $P^{Har}_{ ext{Norm}}$ and bi-level updates for fast convergence. Simulation results show that STAR-RIS with Meta-DDPG outperforms conventional RIS and standard DDPG in terms of convergence speed and the rate-energy trade-off, closely approaching the performance of exhaustive search under given constraints, with tunable performance via the weight $\alpha$. The work demonstrates the practical potential of intelligent surfaces and meta-learning for efficient resource allocation in next-generation wireless networks.

Abstract

Simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) is a cutting-edge concept for the sixth-generation (6G) wireless networks. In this paper, we propose a novel system that incorporates STAR-RIS with simultaneous wireless information and power transfer (SWIPT) using rate splitting multiple access (RSMA). The proposed system facilitates communication from a multi-antenna base station (BS) to single-antenna users in a downlink transmission. The BS concurrently sends energy and information signals to multiple energy harvesting receivers (EHRs) and information data receivers (IDRs) with the support of a deployed STAR-RIS. Furthermore, an optimization is introduced to strike a balance between users' sum rate and the total harvested energy. To achieve this, an optimization problem is formulated to optimize the energy/information beamforming vectors at the BS, the phase shifts at the STAR-RIS, and the common message rate. Subsequently, we employ a meta deep deterministic policy gradient (Meta-DDPG) approach to solve the complex problem. Simulation results validate that the proposed algorithm significantly enhances both data rate and harvested energy in comparison to conventional DDPG.

Figures (5)

• Figure 1: System model of an STAR-RIS SWIPT system with RSMA.
• Figure 2: Convergence of the Meta-DDPG-based algorithm with $\alpha=0.5$.
• Figure 3: The objective versus maximum transmit power with $\alpha=0.5$.
• Figure 4: Data rate versus ANHE.
• Figure 5: The ANHE versus $\alpha$.