Deterministic transmission in user-scalable cell-free MIMO-OFDM systems

Abstract

This paper proposes a data- and model-driven collaborative resource scheduling method to maximize the spectral efficiency (SE) of cell-free (CF) downlink multiuser multiple-input multiple-output (MIMO) systems, subject to delay violation probability and power constraints. The method integrates the weighted minimum mean square error (WMMSE) algorithm within the safety reinforcement learning (Safety-RL) framework. The original optimization problem is decomposed into two coupled subproblems. The Safety-RL algorithm leverages state features to determine user priority weights and allocate bandwidths, while the WMMSE algorithm calculates the precoding matrix and further schedules resources based on user priority weights to obtain the reward and costs of Safety-RL. Considering dynamic user access in CF systems, a distributed algorithm with user scalability is also proposed. Simulation results demonstrate that the proposed approach improves the SE while meeting the different delay violation probability constraints of users. Furthermore, the distributed algorithm offers comparable performance to the fully centralized method while considerably reducing model training overhead, particularly as users dynamically access the system.

Keywords

delay violation probability constraint / cell-free / safety reinforcement learning / resource scheduling

References

[1] ZHANG H, CHEN L Q, YANG B, et al. Secure lightweight data using scheme in 5G industrial Internet systems[J]. Journal of Southeast University (Natural Science Edition), 2024, 54(3): 772-780. (in Chinese)

[2] JIANG W, SONG T C, WANG C, et al. Resource allocation for NOMA-based vehicular communications[J]. Journal of southeast University (English Edition), 2020, 36(1): 8-13.

[3] SONG X Q, ZHANG W J, LEI L, et al. Dynamic resource allocation algorithm for multi-objective joint optimization in Internet of vehicle[J]. Journal of Southeast University (Natural Science Edition), 2025, 55(1): 266-274. (in Chinese)

[4] LU Y Z, YANG L, YANG S X, et al. An intelligent deterministic scheduling method for ultralow latency communication in edge enabled industrial Internet of Things[J]. IEEE Transactions on Industrial Informatics, 2022, 19(2): 1756-1767.

[5] POLYANSKIY Y, POOR H V, VERDÚ S. Channel coding rate in the finite blocklength regime[J]. IEEE Transactions on Information Theory, 2010, 56(5): 2307-2359.

[6] TOMASI B, PREISIG J C. Energy-efficient transmission strategies for delay constrained traffic with limited feedback[J]. IEEE Transactions on Wireless Communications, 2014, 14(3): 1369-1379.

[7] LU B B, ZHANG H X, XUE T, et al. Deep reinforcement learning-based power allocation for ultra reliable low latency communications in vehicular networks[C]// 2021 IEEE/CIC International Conference on Communications in China (ICCC). Xiamen, China, 2021: 1149-1154.

[8] HU F J, ZHAO J H, LIAO J Y, et al. Deep reinforcement learning based resource allocation for URLLC user-centric network[C]// 2022 14th International Conference on Wireless Communications and Signal Processing (WCSP). Nanjing, China, 2022: 522-526.

[9] TIAN C, HUANG G, LIU A, et al. Delay-aware power control for downlink multi-user MIMO via constrained deep reinforcement learning[C]// 2021 IEEE Global Communications Conference (GLOBECOM). Madrid, Spain, 2021: 1-6.

[10] SHAFIEIRAD H, ADVE R S. On meeting a maximum delay constraint using reinforcement learning[J]. IEEE Access, 2022, 10: 97897-97911.

[11] ZHOU F F, ZHENG J, YANG L Y R, et al. A DQN-based joint spectrum and computing resource allocation algorithm for multi-service MEC networks[C]// ICC 2022—IEEE International Conference on Communications. Seoul, Republic of Korea, 2022: 5663-5669.

[12] YANG L, JI J M, DAI J T, et al. Constrained update projection approach to safe policy optimization[J]. Advances in Neural Information Processing Systems, 2022, 35: 9111-9124.

[13] FENG C H, SHEN W Q, AN J P, et al. Weighted sum rate maximization of the mmWave cell-free MIMO downlink relying on hybrid precoding[J]. IEEE Transactions on Wireless Communications, 2021, 21(4): 2547-2560.

Funding

National Natural Science Foundation of China(62271140)

National Natural Science Foundation of China(62225107)

Natural Science Foundation of Jiangsu Province(BK20240174)

Fundamental Research Funds for the Central Universities(2242022k60002)

Fund of Jiangsu Provincial Scientific Research Center of Applied Mathematics(BK20233002)