Overcoming the Phase Discontinuity Problem in RIS Optimization: A Robust and Scalable Phase-Aware Deep Regression Framework via Hybrid CNN–LSTM Architecture
Advanced Intelligent Systems, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Basım Tarihi: 2026
- Doi Numarası: 10.1002/aisy.70460
- Dergi Adı: Advanced Intelligent Systems
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Applied Science & Technology Source, Compendex, INSPEC, Directory of Open Access Journals, Technology Collection (ProQuest)
- Anahtar Kelimeler: 6G wireless networks, beamforming, circular manifold regression, deep regression, hybrid CNN–LSTM architecture, phase discontinuity, reconfigurable intelligent surfaces
- Bursa Uludağ Üniversitesi Adresli: Evet
Özet
Reconfigurable intelligent surfaces (RIS) are crucial for 6G connectivity. However, data-driven RIS optimizers suffer from the phase discontinuity problem, where estimation collapses near (Formula presented.) boundaries due to topological mismatches between the circular phase manifold (Formula presented.) and Euclidean output spaces. To overcome this, we propose a robust, phase-aware hybrid CNN–LSTM framework that maps the cyclic phase-space onto continuous Euclidean coordinates via orthogonal sine–cosine projections. Supported by formal (Formula presented.) embedding proofs, the framework is evaluated on 40,000 channel realizations. It achieves a (Formula presented.) root-mean-square error, outperforming established benchmarks by 21%. It yields a 0.08 ms inference latency ((Formula presented.) faster than particle swarm optimization). With (Formula presented.) linear complexity, it maintains a 0.64 ms delay for (Formula presented.) elements, meeting URLLC requirements. Ablation studies isolate 29.9% and 20.9% accuracy gains from sine–cosine encoding and LSTM spatial recurrence, respectively. Robustness tests confirm a sustained 15% margin over classical solvers under 10% channel estimation impairment. This scalable architecture offers a superior alternative to quadratic-complexity ((Formula presented.)) Transformer models, representing one of the first principled deployments of sine–cosine encoding for RIS optimization.