Joint Active and Passive Beamforming Optimization for Beyond Diagonal RIS-aided Multi-User Communications

2025-01-17

IEEE Communications Letters

AI Summary

This work addresses the high computational complexity and performance trade-off in joint active/passive beamforming optimization for BD-RIS-aided multi-user MIMO systems under symmetric unitary constraints. To tackle this, we propose an alternating iterative closed-form algorithm that integrates matrix decomposition, the weighted minimum mean square error (WMMSE) framework, and gradient projection. Furthermore, we introduce a beyond-diagonal structural model for the BD-RIS and a joint channel reconstruction technique, thereby overcoming fundamental limitations of conventional passive beamforming. The proposed method achieves near-optimal weighted sum-rate maximization with low computational overhead: it reduces complexity by over 60% compared to state-of-the-art approaches, significantly outperforms heuristic and semidefinite relaxation-based benchmarks in rate performance, and closely approaches the optimal solution.

Technology Category

• Search and Optimization: Non-convex Optimization
• Machine Learning: Optimization
• Constraint Satisfaction and Optimization: Mixed Discrete/Continuous Optimization

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Benefiting from its capability to generalize existing reconfigurable intelligent surface (RIS) architectures and provide additional design flexibility via interactions between RIS elements, beyond-diagonal RIS (BD-RIS) has attracted considerable research interests recently. However, due to the symmetric and unitary passive beamforming constraint imposed on BD-RIS, existing joint active and passive beamforming optimization algorithms for BD-RIS either exhibit high computational complexity to achieve near optimal solutions or rely on heuristic algorithms with substantial performance loss. In this paper, we address this issue by proposing an efficient optimization framework for BD-RIS assisted multi-user multi-antenna communication networks. Specifically, we solve the weighted sum rate maximization problem by introducing a novel beamforming optimization algorithm that alternately optimizes active and passive beamforming matrices using iterative closed-form solutions. Numerical results demonstrate that our algorithm significantly reduces computational complexity while ensuring a sub-optimal solution.

Problem

• Multi-User Multi-Antenna Communication
• BD-RIS Beamforming Optimization
• Efficiency and Complexity Reduction

Innovation

• BD-RIS Optimization
• Efficient Algorithm
• Multi-User Multi-Antenna Networks