Effective capacity analysis of cognitive radio channels for quality of service provisioning

In this paper, the performance of cognitive radio systems is studied when the secondary users operate under statistical quality of service (QoS) constraints. In the cognitive radio channel model, secondary users initially perform channel sensing, and then engage in data transmission at two different average power levels depending on the channel sensing results. A state transition model is constructed to model this cognitive transmission channel. Statistical QoS constraints are imposed as limitations on buffer violation probabilities. Effective capacity of the cognitive radio channel, which provides the maximum throughput under such QoS constraints, is determined. This analysis is conducted for fixed-power/fixed-rate, fixed-power/variable-rate, and variable-power/variable-rate transmission schemes under different assumptions on the availability of channel side information (CSI) at the transmitter. The interactions and tradeoffs between the throughput, QoS constraints, and channel sensing parameters (e.g., sensing duration and threshold, and detection and false alarm probabilities) are investigated. The performances of fixed-rate and variable-rate transmission methods are compared in the presence of QoS limitations. It is shown that variable schemes outperform fixed-rate transmission techniques if the detection probabilities are high. Performance gains through adapting the power and rate are quantified and it is shown that these gains diminish as the QoS limitations become more stringent.