Scaling laws of cognitive networks with cooperative relay

This study deals with the scaling laws of ad hoc cognitive networks. In the network, m primary users and n secondary users (m<n) are randomly located in the same geographic area and share one common frequency spectrum. By loosening some of the restrictions for the network, the authors obtained the throughput upper bounds for both primary and secondary users. To obtain the lower bounds (achievable throughputs), we proposed a cooperative transmission scheme between the primary and secondary users by using the percolation theory, and the results showed that the achievable throughputs can meet the upper bounds, thus, the authors obtained the throughputs. On the basis of our results, per primary user can transmit with a throughput of min(Θ(1), Θ(√n/m)), which is higher than the throughput of Θ(√1/m) in the cognitive network without cooperative relay. This may serve as a strong incentive for primary users to cooperative with secondary users. Meanwhile, secondary users can achieve the same scale as if there is no primary network exists.