Loading [MathJax]/extensions/MathMenu.js
Generative Diffusion Model-Based Deep Reinforcement Learning for Uplink Rate-Splitting Multiple Access in LEO Satellite Networks | IEEE Conference Publication | IEEE Xplore

Generative Diffusion Model-Based Deep Reinforcement Learning for Uplink Rate-Splitting Multiple Access in LEO Satellite 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,...Show More

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.
Date of Conference: 26-29 June 2024
Date Added to IEEE Xplore: 31 October 2024
ISBN Information:

ISSN Information:

Conference Location: Paris, France

Funding Agency:


I. Introduction

To ensure the ubiquitous global coverage, the amalgamation of satellites and traditional networks is imperative for 5G-Advanced and even 6G systems [1], [2]. In particular, low earth orbit (LEO) satellites can provide lower service latency for delay-sensitive applications due to its lower orbital altitude, and as well higher data rate for bandwidth-sensitive services due to its larger constellation density [3]. Further, LEO satellites can improve the backhaul capability in base stations (BSs), and work as access points to offload tasks of ground terminals with worse channel conditions to BSs [4].

References

References is not available for this document.