Energy-efficient user association in cognitive heterogeneous networks

Due to the ever increasing data traffic demands, which are directly connected to increased energy consumption, it becomes challenging for operators to achieve capacity enhancement while limiting their electric bill. To that end, exploiting the context awareness of future cognitive networks is expected to play a key role. Next generation cellular networks are about to include a plethora of small cells, with users being able to communicate via multiple bands. Given that small cells are expected to be eventually as close as 50 m apart, not all of them will have a direct connection to the core network; thus, multihop communication through neighboring small cells may be required. In such architectures, the user association problem becomes challenging, with backhaul energy consumption being a definitive parameter. Thus, in this article, we study the user association problem in cognitive heterogeneous networks. We evaluate the existing approaches in terms of energy efficiency and show the potential of exploiting the available context-aware information (i.e., users' measurements and requirements, knowledge of the network architecture, and the available spectrum resources of each base station) to associate the users in an energy-efficient way, while maintaining high spectrum efficiency. Our study considers both the access network and backhaul energy consumption, while the performance of the association algorithms is evaluated under two different case study scenarios.Peer Reviewe