A Review on Sum-Rate Maximization Techniques in SIMO-RSMA System with DRL Approach: Research Challenges and Future Trends

  • Mac, Huy Dang; 
  • Nguyen Tuan Tran, Kiet; 
  • Won, Dongwook; 
  • Cho, Sungrae
Citations

SCOPUS

0

초록

Rate-Splitting Multiple Access (RSMA) has emerged as a powerful multiple access technique for interference management in multi-user wireless networks. This paper presents a comprehensive review of sum-rate maximization techniques in a Single-Input Multiple-Output (SIMO) RSMA system, with a particular focus on approaches leveraging Deep Reinforcement Learning (DRL). We discuss how DRL methods are being employed to tackle the challenging joint optimization of transmit power, beamforming, and decoding order in RSMA systems to maximize network throughput. The Abstract highlights recent advances in applying DRL for sum-rate maximization in SIMO-RSMA - such as Deep Deterministic Policy Gradient and Soft Actor-Critic algorithms - and emphasizes their performance gains in dynamic environments. Key research challenges (e.g., computational complexity, convergence, partial observability) are identified, and future trends (including integration with reconfigurable intelligent surfaces, meta-learning, and network slicing) are outlined. This survey provides researchers and practitioners with insights into state-of-the-art RSMA techniques and DRL-driven solutions for next-generation wireless networks.

키워드

6G; deep reinforcement learning; O-RAN; RIS; RSMA; SIMO; sum-rate maximization
제목
A Review on Sum-Rate Maximization Techniques in SIMO-RSMA System with DRL Approach: Research Challenges and Future Trends
저자
Mac, Huy Dang; Nguyen Tuan Tran, Kiet; Won, Dongwook; Cho, Sungrae
DOI
10.1109/ICTC66702.2025.11388705
발행일
2025
유형
Conference Paper
저널명
International Conference on ICT Convergence
페이지
257 ~ 262