Clustered cell-free networking paves a new way for enabling scalable joint transmission among access points (APs) by partitioning the whole network into non-overlapping sub-networks. Previous works adopted clustering algorithms, graph partitioning methods or conventional continuous optimization theories to partition a network based on the channels between all users and all APs, resulting in huge channel measurement and computational costs. This makes these methods difficult to be implemented in practical systems since the optimal network partition could vary frequently due to user mobility. In addition, existing methods were usually designed for specific clustered cell-free networking problems with different optimization algorithms employed. In this paper, we leverage deep reinforcement learning (DRL) for clustered cell-free networking so as to rapidly adapt to user movements in dynamic environments, and propose a deep deterministic policy gradient based clustered cell-free networking (DDPG-C2F) framework that can be adapted in various application scenarios. Moreover, in our framework, only one single channel needs to be estimated at each AP as the input of the neural network, which greatly reduces the channel measurement costs for clustered cell-free networking, and the training and inference costs of our framework. The proposed DDPG-C2F framework is then applied to various clustered cell-free networking problems with different objectives and constraints to demonstrate its performance. Simulation results show that our framework outperforms existing baselines in all scenarios. Moreover, we show that the proposed framework can reduce the handover cost over user mobility, and is robust to dynamic scenarios with random user joining or leaving.

Clustered cell-free networking paves a new way for enabling joint transmission among access points (APs) by decomposing the whole network into non-overlapping subnetworks. Previous works adopted clustering algorithms or conventional optimization theories to group users and APs into subnetworks, which usually require high computational costs and have long running time. In this paper, we leverage deep reinforcement learning (DRL) for clustered cell-free networking so as to rapidly adapt to user movement in dynamic scenarios. To effectively reduce the joint processing complexity of each subnetwork, we aim to maximize the balance of the subnetworks in addition to the sum rate. With such an objective, we propose a novel deep deterministic policy gradient based clustered cell-free networking (DDPG-C2F) framework. It is worth mentioning that the proposed framework requires much lower channel measurement overhead and computational complexity compared to the conventional approaches. Simulation results demonstrate the superior performance of the proposed DDPG-C2F framework in various scenarios.

Clustered cell-free networking that dynamically decomposes the whole network into a number of subnetworks operating in parallel could help avoid the severe cell-edge problem. In this paper, the clustered cell-free networking problem is investigated with access point (AP) selection considered for power saving. A novel user-centric ratio-fixed AP-selection based clustering (UCR-ApSel) algorithm is proposed. Simulation results show that the proposed UCR-ApSel algorithm improves not only the rate performance but also user fairness. The average per-user rate achieved with the proposed UCR-ApSel algorithm is further analyzed and a closed-form upper-bound is derived. Simulation results show that the gap between the derived upper-bound and the average per-user rate remains almost a constant as the system parameters vary, indicating that the closed-form upper-bound could provide direct guidance to practical system design and performance optimization.