Joint Cell Activation and User Association for Backhaul Load Balancing in Green HetNets

This letter proposes a joint cell activation and user association scheme to address the power consumption and backhaul load balancing issues in green dense heterogeneous networks (HetNets). In this scheme, an optimization problem is formulated for jointly minimizing power consumption and optimizing backhaul load balance of dense HetNets, subject to diverse quality of service (QoS) requirements and heterogeneous backhaul capacities. To solve this problem, a Q-learning-based algorithm is developed. Results show that the proposed scheme achieves substantial performance improvements over several baseline schemes in terms of fairness, QoS and energy efficiency.