Delay optimal cognitive multiple access: a superposition coding approach

Cognitive radio (CR) is a powerful solution that can significantly improve the radio spectrum resources efficiency by allowing secondary users (SUs) to borrow unused spectrum from primary licensed networks. Conventional CR protocols allow an SU to transmit only when the primary users (PUs) are not active. However, waiting for idle timeslots may induce large packet delay and loss, so as to result in poor quality of service (QoS) for the SU. To overcome this, a joint coding and scheduling (JCS) method for cognitive multiple access is proposed. In particular, a successive interference decoder is utilized in the physical layer to thoroughly mitigate the SU’s interference to PU. A joint channel and queue aware scheduling is then proposed for the MAC layer of SU to minimize its average packet delay given a power constraint. We shall adopt Markov models to derive the analytical results on delay, packet-loss rate, and power consumption of the proposed scheme. Then by formulating and solving cross-layer optimization problem, we obtain the optimal scheduling parameter and the performance limit of the JCS in terms of delay-power tradeoff.