Energy-aware activation of nomadic relays for performance enhancement in cellular networks

This paper presents an optimization framework for energy-aware relay selection and user association in cellular networks aided by nomadic relays. The framework of sparse optimization is used to minimize network energy consumption for desired service provisioning of the terminals. We show that some constraints in the underlying optimization are of quadratic form due to the assumption of relays with wireless backhaul links. Hence, previously proposed algorithms for activation of network elements with wired backhaul links are not applicable. In this paper, therefore, novel algorithms based on different relaxations of the quadratic constraints are proposed and evaluated for energy savings. Simulation results confirm that the proposed algorithms may significantly reduce the overall energy consumption of cellular networks compared with conventional cell selection schemes.