PPO-Based Adaptive Beamforming for Single-User MISO Systems in 6G Networks
Résumé
Sixth-generation (6G) wireless networks are expected to operate under highly dynamic channel conditions while supporting stringent performance requirements in terms of spectral efficiency and reliability. Beamforming plays a critical role in enabling these capabilities. Conventional beamforming techniques rely on accurate channel state information and analytical optimization assumptions, which may limit their adaptability in rapidly varying environments. This paper investigates an adaptive beamforming framework based on Proximal Policy Optimization (PPO) for single-user Multiple-Input Single-Output (MISO) systems. The beamforming task is formulated as a continuous-control reinforcement learning problem in which the learning agent adjusts the beamforming vector through interaction with the wireless environment. The proposed framework is evaluated in a controlled single-user MISO scenario under Rayleigh fading channels and compared with classical beamforming methods such as Maximum Ratio Transmission (MRT), Zero-Forcing (ZF), and Minimum Mean Square Error (MMSE) precoding. Performance is analyzed using training convergence behavior, instantaneous and average spectral efficiency, cumulative distribution functions of spectral efficiency and SINR, and beam pattern visualization. Simulation results show that the PPO-based approach learns stable beamforming policies and achieves competitive spectral efficiency while maintaining consistent learning behavior across stochastic channel realizations. These results demonstrate that reinforcement learning can provide a transparent and reproducible framework for studying adaptive beamforming strategies and offer insights into the learning dynamics of data-driven physical-layer optimization for future wireless communication systems.
Citer ce document
Exporter : BibTeX · RIS (Zotero, Mendeley, EndNote)
Accès au document
Texte intégral archivé sur SPHAERO, en accès ouvert : lisible gratuitement en ligne après connexion. Le document reste aussi disponible à sa source. Se connecter
Voir l'article sur le site de la revueLicence et provenance
Licence : CC BY
Notice moissonnée depuis OpenAlex le 12/10/2026. Le document reste hébergé par sa source.
Voir le document à la source →