Bounded Deep Unfolding for Joint Beamforming and Scheduling in Multi-Cell MIMO Networks

2026-06-03

AI Summary

This work addresses the joint resource block group scheduling and beamforming optimization problem in multi-cell MIMO downlink networks by proposing a deep unfolding framework that integrates algorithmic priors. The framework consists of two components: P-Net and K-Net. P-Net accelerates continuous beamforming optimization and ensures convergence by incorporating a bounded adaptive relaxation factor inspired by FastFP, while K-Net employs a long-horizon priority strategy to guide low-complexity greedy scheduling, thereby avoiding the high computational overhead of Hungarian matching. Leveraging recursive parameter sharing and greedy assignment, the proposed method generalizes across varying network scales, antenna configurations, and channel conditions without retraining. It achieves significant improvements in weighted sum rate and substantially reduces execution time compared to conventional model-driven approaches.

Abstract

This paper investigates the joint resource block group (RBG) scheduling and beamforming optimization problem for weighted sum-rate (WSR) maximization in multi-cell multiple-input multiple-output (MIMO) downlink networks. While the Fast Fractional Programming (FastFP) framework provides a reliable model-driven solution, it suffers from conservative continuous beamforming updates and prohibitive computational overhead during the discrete RBG matching phase. To address these bottlenecks, we propose a joint deep unfolding framework comprising two core modules: P-Net and K-Net. For continuous beamforming, P-Net learns an adaptive relaxation factor along the analytical FastFP update direction. By strictly constraining this factor within an ascent-preserving interval, P-Net accelerates the optimization trajectory while rigorously retaining monotonic improvement and stationary-point convergence guarantees. For discrete RBG scheduling, K-Net learns a long-horizon priority policy that guides a low-complexity greedy assignment, effectively preserving the assignment quality while bypassing the high complexity of Hungarian matching. Both networks leverage analytical algorithmic priors and utilize recurrent parameter sharing, enabling flexible inference beyond the training horizon. Extensive simulations demonstrate that the proposed joint framework achieves higher WSR and faster execution times than conventional model-driven baselines, while generalizing robustly across unseen network scales, antenna configurations, and channel conditions without retraining.

Problem

• beamforming
• scheduling
• MIMO networks
• weighted sum-rate maximization
• resource block group

Innovation

• Deep Unfolding
• Beamforming
• Resource Scheduling
• Monotonic Convergence
• Algorithm Unrolling