MaxDiv: an optimal randomized spectrum access with maximum diversity scheme for cognitive radio networks

Abstract

Cognitive radio network (CRN) is an emerging technology that can increase the utilization of spectrum underutilized by primary users (PUs). In the literature, most exiting investigations on CRNs have focused on how secondary users (SUs) can coexist harmlessly with the PUs. Despite the importance of such a coexistence issue, it is also crucial to investigate the coexistence of SUs because (i) the PUs usually rarely use the licensed spectrum and (ii) the advantages of CRN will significantly increase the number of SUs in the future. To address this challenging issue, we propose, in this paper, an optimal randomized spectrum access scheme, whose main ideas include the following: (i) an SU shares its sensing results with neighboring SUs and (ii) with the regional sensing results, an SU will access available channels with a non-uniform probability distribution. We first formulate a multichannel optimal randomized multiple access (MC-ORMA) problem that aims to maximize the throughput of the CRN; we then develop efficient distributed algorithms to solve the MC-ORMA problem; we derive the closed-form value of collision probability for each SU; and finally, we conduct extensive numerical experiments and compare our theoretical analysis with simulation results to demonstrate the advantages of our scheme. Copyright © 2011 John Wiley & Sons, Ltd.