Abstract

The high spectrum efficiency of massive multiple input multiple output (mMIMO)-enabled beamforming is attractive for solving the massive number of connections in next-generation radio access networks (NG-RANs). However, the beamforming based on multiple antenna sub-arrays to support multi-connections would introduce extra 2D antenna sub-array selection and radio resource block (RB) allocation; meanwhile, considering the front-haul bandwidth consumption would reduce as much as possible the redundant data transmitting over front-haul in the NG-RANs. Therefore, a key issue is how to coordinate 2D antenna sub-array selection and radio RB allocation to minimize the front-haul bandwidth and maximize the radio RB utilization when accommodating a set of beam antenna array (BAA) requests. In this paper, we first establish a 3D BAA mapping model to jointly optimize 2D antenna sub-array selection and radio RB allocation, which would in turn affect the allocation of the front-haul wavelength bandwidth in a time and wavelength division multiplexed passive optical network (TWDM-PON)-based front-haul with the mMIMO antenna system. An integer linear programming mathematical model is formulated, and a novel deep reinforcement learning (DRL)-based algorithm is proposed to optimize the front-haul bandwidth and radio RB utilization. Besides, two heuristic algorithms are developed with different antenna sub-array selection policies as the benchmarks. The extensive simulation results show that our proposed DRL-based algorithm can attain the lowest average cost (AC) for different numbers of BAA requests considered by seeking the optimal trade-off between the front-haul bandwidth consumption and the number of the used antennas. Compared with the inter-sub-array benchmark, the DRL-based algorithm can achieve up to 7.5% AC reduction, which is mainly attributed to the 21.2% reduction of the front-haul bandwidth without increasing the number of used antennas.

© 2023 Optica Publishing Group

Security-aware 5G RAN slice mapping with tiered isolation in physical-layer secured metro-aggregation elastic optical networks using heuristic-assisted DRL

Yunwu Wang, Min Zhu, Jiahua Gu, Xiang Liu, Weidong Tong, Bingchang Hua, Mingzheng Lei, Yuancheng Cai, and Jiao Zhang
J. Opt. Commun. Netw. 15(12) 969-984 (2023)

Resource-efficient and QoS guaranteed 5G RAN slice migration in elastic metro aggregation networks using heuristic-assisted deep reinforcement learning

Jiahua Gu, Min Zhu, Yunwu Wang, Xiaofeng Cai, Yuancheng Cai, Jiao Zhang, Mingzheng Lei, Bingchang Hua, Pingping Gu, and Guo Zhao
J. Opt. Commun. Netw. 15(11) 854-870 (2023)

Performance Evaluation of XG-PON Based Mobile Front-Haul Transport in Cloud-RAN Architecture

Ahmed Mohammed Mikaeil, Weisheng Hu, Tong Ye, and Syed Baqar Hussain
J. Opt. Commun. Netw. 9(11) 984-994 (2017)

Dynamic IQ Data Compression Using Wireless Resource Allocation for Mobile Front-Haul With TDM–PON [Invited]

Naotaka Shibata, Shigeru Kuwano, Jun Terada, and Hideaki Kimura
J. Opt. Commun. Netw. 7(3) A372-A378 (2015)

Edge-enhanced graph neural network for DU-CU placement and lightpath provision in X-Haul networks

Ruikun Wang, Jiawei Zhang, Zhiqun Gu, Shuangyi Yan, Yuming Xiao, and Yuefeng Ji
J. Opt. Commun. Netw. 14(10) 828-839 (2022)