A QoS-aware framework for available spectrum characterization and decision in Cognitive Radio networks

Abstract

The growing problem of spectrum scarcity and the inefficient spectrum utilization in the licensed bands, are addressed by the emerging Cognitive Radio (CR) paradigm. It is seen that the choice of the spectrum bands, called as spectrum decision, must be organized carefully by considering the challenges in the spectrum availability over time, the short term fluctuations in the availability, and the heterogeneous Quality of Service (QoS) requirements of the cognitive radio users. Taking into account these challenges, the main contribution of this paper is to design a QoS-aware spectrum decision framework that achieves higher throughput and fairness in CR networks. The available spectrum fluctuations are characterized by using queueing theoretic models and are parametrized by a novel QoS parameter called opportunity index, Ψ. The heterogeneous QoS requirements of CR users are classified by defining another novel QoS parameter called request index, κ. An admission control algorithm is designed to stabilize the heterogeneous QoS requirements according to Ψ and κ. A spectrum decision algorithm is developed to select most appropriate spectrum bands considering the stabilized QoS requirements and the characterized spectrum. Moreover, by using a novel spectrum mobility algorithm the available spectrum is continuously monitored for dynamic variations in the CR network. The simulations demonstrate that our QoS-aware spectrum characterization and decision framework maximizes the total throughput while maintaining the fairness.