Simultaneous sense and access for cognitive radio networks: A throughput optimization perspective

In this paper, we propose a novel frame structure for spectrum sensing in a network of secondary users, and optimize the sum throughput of the network under the constraint on the aggregate interference offered by the network to the primary user. We contrast its performance with the traditional Stop And Sense (SAS) frame structure, which involves the secondary users stopping transmissions and sensing for spectrum holes. We obtain near-optimal, easily computable solutions for the sensing duration and sensing thresholds that approximately maximize the secondary throughput under an aggregate interference constraint. Through Monte Carlo simulations, we present a detailed comparison of the two schemes, and characterize the operating regimes where each frame structure is preferable over the other. In general when switching latencies are not accounted for, we find that the proposed Simultaneous Sense and Access (SSA) frame structure performs similar to the existing SAS frame structure in terms of throughput for all values of primary signal SNR. When switching latencies are accounted for, the SSA frame structure offers significantly higher throughput gain for all values of primary SNR. Another advantage of this frame structure over a few other existing frame structures when employed in a network of secondary users is that it does not rely on the data in the previous slot to make decisions about the presence/absence of primary signal in the current slot which makes the frame structure ideal for an adhoc network.