QoS-aware cross-layer scheduling for cognitive radio networks with heterogeneous data traffic

Cognitive radio (CR) has been an enabling technology for secondary users (SUs) to efficiently exploit radio spectrum underutilized by primary users (PUs). To avoid downgrading PUs' performance, dynamic spectrum access (DSA) plays an important role in CR networks (CRNs) for adaptively assigning spectrum to SUs. During spectrum handover to PUs, quality of service (QoS) of SUs may not be satisfied due to high competition among SUs with different traffic types. However, in most existing DSA schemes, traffic types handled by SUs are treated equally. To achieve QoS satisfaction for SUs with heterogeneous traffic, this paper proposes a new cross-layer scheduling that jointly considers DSA, packet scheduling and a channel sharing policy. QoS performance of the cross-layer scheduling is measured in terms of average throughput, average packet delay and dropping probability for SUs with different traffic types. QoS support for SUs provided by the proposed scheduling scheme incorporating the channel sharing policy is shown to be more efficient and balanced compared to existing scheduling schemes.