2020/01/28 by Yasser Al-Eryani, Mohamed Akrout, Al-Eryani, Yasser +3 · 1 citation
Engineering · #Advanced MIMO Systems Optimization #Antenna Design and Analysis #Energy Harvesting in Wireless Networks #FOS: Computer and information sciences #FOS: Electrical engineering #Full-Duplex Wireless Communications #Information Theory (cs.IT) #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Signal Processing (eess.SP) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2002.02801
openalex publication_date 2020/01/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In future cell-free (or cell-less) wireless networks, a large number of\ndevices in a geographical area will be served simultaneously in non-orthogonal\nmultiple access scenarios by a large number of distributed access points (APs),\nwhich coordinate with a centralized processing pool. For such a centralized\ncell-free network with static predefined beamforming design, we first derive a\nclosed-form expression of the uplink per-user probability of outage. To\nsignificantly reduce the complexity of joint processing of users' signals in\npresence of a large number of devices and APs, we propose a novel dynamic\ncell-free network architecture. In this architecture, the distributed APs are\npartitioned (i.e. clustered) among a set of subgroups with each subgroup acting\nas a virtual AP equipped with a distributed antenna system (DAS). The\nconventional static cell-free network is a special case of this dynamic\ncell-free network when the cluster size is one. For this dynamic cell-free\nnetwork, we propose a successive interference cancellation (SIC)-enabled signal\ndetection method and an inter-user-interference (IUI)-aware DAS's receive\ndiversity combining scheme. We then formulate the general problem of clustering\nAPs and designing the beamforming vectors with an objective to maximizing the\nsum rate or maximizing the minimum rate. To this end, we propose a hybrid deep\nreinforcement learning (DRL) model, namely, a deep deterministic policy\ngradient (DDPG)-deep double Q-network (DDQN) model, to solve the optimization\nproblem for online implementation with low complexity. The DRL model for\nsum-rate optimization significantly outperforms that for maximizing the minimum\nrate in terms of average per-user rate performance. Also, in our system\nsetting, the proposed DDPG-DDQN scheme is found to achieve around 78 % of the\nrate achievable through an exhaustive search-based design.\n