Beyond-Diagonal RIS Under Non-Idealities: Learning-Based Architecture Discovery and Optimization

2025-10-17

AI Summary

Designing beyond-diagonal reconfigurable intelligent surfaces (BD-RIS) under non-ideal circuit conditions poses significant challenges in balancing architectural feasibility, performance, and hardware complexity. Method: This paper proposes a learning-driven bi-level joint optimization framework: an upper level employs a deep generative model to efficiently sample candidate architectures from a large design space; the lower level integrates non-ideal circuit modeling and differentiable performance evaluation, enabling end-to-end trainable co-optimization of architecture, performance, and complexity. Contribution/Results: Our approach is the first to systematically characterize how circuit non-idealities influence BD-RIS architecture selection, circumventing local optima inherent in conventional heuristic search. Experiments under diverse hardware complexity constraints demonstrate that the discovered architectures achieve substantial gains—average +4.2 dB improvement in signal-to-interference-plus-noise ratio (SINR) and +38% enhancement in energy efficiency—establishing a scalable, verifiable, and automated design paradigm for practical BD-RIS deployment.

Technology Category

• Search and Optimization: Learning to Search
• Intelligent Robots: Learning & Optimization for ROB
• Computer Vision: Learning & Optimization for CV

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• 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

Abstract

Beyond-diagonal reconfigurable intelligent surface (BD-RIS) has recently been introduced to enable advanced control over electromagnetic waves to further increase the benefits of traditional RIS in enhancing signal quality and improving spectral and energy efficiency for next-generation wireless networks. A significant issue in designing and deploying BD-RIS is the tradeoff between its performance and circuit complexity. Despite some efforts in exploring optimal architectures with the lowest circuit complexities for ideal BD-RIS, architecture discovery for non-ideal BD-RIS remains uninvestigated. Therefore, how non-idealities and circuit complexity jointly affect the performance of BD-RIS remains unclear, making it difficult to achieve the performance - circuit complexity tradeoff in the presence of non-idealities. Essentially, architecture discovery for non-ideal BD-RIS faces challenges from both the computational complexity of global architecture search and the difficulty in achieving global optima. To tackle these challenges, we propose a learning-based two-tier architecture discovery framework (LTTADF) consisting of an architecture generator and a performance optimizer to jointly discover optimal architectures of non-ideal BD-RIS given specific circuit complexities, which can effectively explore over a large architecture space while avoiding getting trapped in poor local optima and thus achieving near-optimal solutions for the performance optimization. Numerical results provide valuable insights for deploying non-ideal BD-RIS considering the performance - circuit complexity tradeoff.

Problem

• Optimizing BD-RIS architecture under non-idealities and circuit complexity
• Addressing performance-circuit complexity tradeoff in non-ideal BD-RIS deployment
• Solving computational challenges in global architecture discovery and optimization

Innovation

• Learning-based framework discovers non-ideal BD-RIS architectures
• Two-tier approach jointly optimizes performance and complexity
• Avoids local optima to achieve near-optimal electromagnetic control