Scalable AP Clustering With Deep Reinforcement Learning for Cell-Free Massive MIMO

被引:0
作者
Tsukamoto, Yu [1 ]
Ikami, Akio [1 ]
Murakami, Takahide [1 ]
Amrallah, Amr [1 ]
Shinbo, Hiroyuki [1 ]
Amano, Yoshiaki [1 ]
机构
[1] KDDI Res Inc, Adv Radio Applicat Lab, Saitama 3568502, Japan
来源
IEEE OPEN JOURNAL OF THE COMMUNICATIONS SOCIETY | 2025年 / 6卷
关键词
Load modeling; Computational modeling; Throughput; Signal processing; Computer architecture; Channel estimation; Signal to noise ratio; Interference; Numerical models; Massive MIMO; Cell-free massive MIMO; access point clustering; deep reinforcement learning; COORDINATED MULTIPOINT TRANSMISSION; DISTRIBUTED ANTENNA SYSTEMS; C-RAN; NETWORK; 5G; ARCHITECTURE; PERFORMANCE; RECEPTION;
D O I
10.1109/OJCOMS.2025.3543681
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Cell-free massive MIMO (CF-mMIMO) is a promising approach for future mobile networks, utilizing centralized MIMO processing for densely distributed access points (APs). In CF-mMIMO, to reduce the computational load for signal processing while meeting throughput demands, user equipment (UEs) are served by a selected number of APs. A significant challenge is AP clustering for each UE, particularly in dynamic environments with moving UEs. One approach for optimizing the AP cluster involves deep reinforcement learning (DRL). However, with numerous UEs and APs, the computational load of DRL increases due to the larger model size and higher inference frequency. To address this, we propose an AP clustering method using distributed DRL. The model focuses on determining the AP cluster for every single UE to prevent model size expansion. The per-user models act as distributed actors, enabling parallel inference. Furthermore, to suppress inference frequency, multiple UEs with low mobility are assigned to the same actor, minimizing the number of parallel actors required without compromising throughput. Numerical simulation shows that our proposed method achieves efficient AP clustering that satisfies throughput requirements with reduced computational load in DRL, even in large-scale environments.
引用
收藏
页码:1552 / 1567
页数:16
相关论文
共 53 条
[1]   A Cooperation Strategy Based on Bargaining Game for Fair User-Centric Clustering in Cloud-RAN [J].
Abdelhakam, Mostafa M. ;
Elmesalawy, Mahmoud M. ;
Mahmoud, Korany R. ;
Ibrahim, Ibrahim. I. .
IEEE COMMUNICATIONS LETTERS, 2018, 22 (07) :1454-1457
[2]   Flexible Resource Block Allocation to Multiple Slices for Radio Access Network Slicing Using Deep Reinforcement Learning [J].
Abiko, Yu ;
Saito, Takato ;
Ikeda, Daizo ;
Ohta, Ken ;
Mizuno, Tadanori ;
Mineno, Hiroshi .
IEEE ACCESS, 2020, 8 :68183-68198
[3]   Multiple Access in Cell-Free Networks: Outage Performance, Dynamic Clustering, and Deep Reinforcement Learning-Based Design [J].
Al-Eryani, Yasser ;
Akrout, Mohamed ;
Hossain, Ekram .
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, 2021, 39 (04) :1028-1042
[4]  
[Anonymous], 2023, ITU-Rec. M.2160-0
[5]   Access Point Clustering in Cell-Free Massive MIMO Using Conventional and Federated Multi-Agent Reinforcement Learning [J].
Banerjee, Bitan ;
Elliott, Robert C. ;
Krzymien, Witold A. ;
Medra, Mostafa .
IEEE TRANSACTIONS ON MACHINE LEARNING IN COMMUNICATIONS AND NETWORKING, 2023, 1 :107-123
[6]  
Biswas P, 2024, IEEE T MACH LEARN CO, V2, P675, DOI [10.1109/tmlcn.2024.3403789, 10.1109/TMLCN.2024.3403789]
[7]   Scalable Cell-Free Massive MIMO Systems [J].
Bjornson, Emil ;
Sanguinetti, Luca .
IEEE TRANSACTIONS ON COMMUNICATIONS, 2020, 68 (07) :4247-4261
[8]   Making Cell-Free Massive MIMO Competitive With MMSE Processing and Centralized Implementation [J].
Bjornson, Emil ;
Sanguinetti, Luca .
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2020, 19 (01) :77-90
[9]   Cell-Free Massive MIMO: User-Centric Approach [J].
Buzzi, Stefano ;
D'Andrea, Carmen .
IEEE WIRELESS COMMUNICATIONS LETTERS, 2017, 6 (06) :706-709
[10]  
Chai X., 2020, P IEEE 91 VEH TECHN, P1