A4 Article in conference proceedings
Generative Diffusion Model-Based Deep Reinforcement Learning for Uplink Rate-Splitting Multiple Access in LEO Satellite Networks (2024)

Wang, X., Wang, K., Zhang, D., Li, J., Zhou, M., & Hämäläinen, T. (2024). Generative Diffusion Model-Based Deep Reinforcement Learning for Uplink Rate-Splitting Multiple Access in LEO Satellite Networks. In 2024 IEEE Symposium on Computers and Communications (ISCC). IEEE Computer Society Press. Proceedings : IEEE Symposium on Computers and Communications. https://doi.org/10.1109/iscc61673.2024.10733704

JYU authors or editors

Publication details

All authors or editors: Wang, Xingjie; Wang, Kan; Zhang, Di; Li, Junhuai; Zhou, Momiao; Hämäläinen, Timo

Parent publication: 2024 IEEE Symposium on Computers and Communications (ISCC)

Place and date of conference: Paris, France, 26.-29.6.2024

ISBN: 979-8-3503-5424-9

eISBN: 979-8-3503-5423-2

Journal or series: Proceedings : IEEE Symposium on Computers and Communications

ISSN: 1530-1346

eISSN: 2642-7389

Publication year: 2024

Publication date: 31/10/2024

Publisher: IEEE Computer Society Press

Publication country: United States

Publication language: English

DOI: https://doi.org/10.1109/iscc61673.2024.10733704

Publication open access: Not open

Publication channel open access: 

Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/98474

Additional information: S05: Non-terrestrial networks

Abstract

This work studies the joint transmit power control and receive beamforming in uplink rate splitting multiple access (RSMA)-based low earth orbit (LEO) satellite networks, using both generative diffusion model and proximal policy optimization (PPO) learning framework. In particular, using RSMA, interference is partially decoded and partially treated as noise, thereby improving the spectral efficiency, while the dynamics and uncertainty in LEO satellite networks would pose challenges to the real-time power control and receive beamforming optimization. First, a long-run sum data rate maximization problem is formulated, subject to the individual data rate requirement, and then the Markov decision process (MDP) is used to model it. Second, on the basis of MDP, a generative diffusion model-based proximal policy optimization (PPO) framework is proposed, where a denoising network is taken as the actor network in PPO to output the optimal continuous policy, thereby facilitating the hyperparameter tuning and improve the sample efficiency. Finally, experiments are conducted to show advantages of merging diffusion model into PPO, in terms of larger spectral efficiency, by comparing proposed framework with benchmarks.

Keywords: data communications networks; communications satellites; satellite communications; wireless data transmission; optimisation; algorithms

Free keywords: LEO satellite networks; RSMA; generative diffusion model; PPO

Contributing organizations

Ministry reporting: Yes

VIRTA submission year: 2024

Preliminary JUFO rating: 1