Optimization and learning in energy efficient resource allocation for cognitive radio networks

Abstract

The recent surge in real-time traffic has led to serious energy efficiency concerns in cognitive radio networks (CRNs). Network infrastructure such as base stations (BSs) host different service classes of traffic with stringent quality-of-service (QoS) requirements that need to be satisfied. Thus, maintaining the desired QoS in an energy efficient manner requires a good trade-off between QoS and energy saving. To deal with this problem, this paper proposes a deep learning-based computational-resource-aware energy consumption technique. The proposed scheme uses an exploration technique of the systems' state-space and traffic load prediction to come up with a better trade-off between QoS and energy saving. The simulation results show that the proposed exploration technique performs 9% better than the traditional random tree technique even when the provisioning priority shifts away from energy saving towards QoS, i.e., α ≥ 0.5.