Energy-efficient Beamforming for RISs-aided Communications: Gradient Based Meta Learning

TL;DR

The paper tackles the non-convex, energy-intensive problem of jointly optimizing active BS precoding and passive RIS phase shifts in RIS-aided MU-MISO networks. It introduces Gradient Based Meta Learning Beamforming (GMLB), which feeds the gradient of the sum-rate into small neural networks (BF-Net and Theta-Net) and uses a differential regulator to stabilize RIS updates, enabling pre-training-free optimization. Through unsupervised meta-learning, GMLB achieves higher sum-rate than traditional alternating-optimization baselines while consuming far less energy, particularly as the RIS size grows. The approach demonstrates strong robustness across SNRs and scales well with system size, offering a practical pathway to energy-efficient, large-scale RIS deployments in 6G and beyond, with future work on incomplete CSI and dynamic scenarios.

Abstract

Reconfigurable intelligent surfaces (RISs) have become a promising technology to meet the requirements of energy efficiency and scalability in future six-generation (6G) communications. However, a significant challenge in RISs-aided communications is the joint optimization of active and passive beamforming at base stations (BSs) and RISs respectively. Specifically, the main difficulty is attributed to the highly non-convex optimization space of beamforming matrices at both BSs and RISs, as well as the diversity and mobility of communication scenarios. To address this, we present a greenly gradient based meta learning beamforming (GMLB) approach. Unlike traditional deep learning based methods which take channel information directly as input, GMLB feeds the gradient of sum rate into neural networks. Coherently, we design a differential regulator to address the phase shift optimization of RISs. Moreover, we use the meta learning to iteratively optimize the beamforming matrices of BSs and RISs. These techniques make the proposed method to work well without requiring energy-consuming pre-training. Simulations show that GMLB could achieve higher sum rate than that of typical alternating optimization algorithms with the energy consumption by two orders of magnitude less.

Figures (7)

• Figure 1: RISs-aided MU-MISO beamforming system.
• Figure 2: Optimization landscapes for RISs-aided communications.
• Figure 3: Workflow of implemented GMLB.
• Figure 4: Performance of the proposed algorithm against the established baselines across varying SNRs.
• Figure 5: Comparison of the proposed algorithm's sum rate with the WMMSE reference over epochs.