Sensing Order Dispersion for Autonomous Cognitive Radios

When autonomous cognitive radios (CRs) have to search multiple potential channels for spectrum opportunities, they face competition from one another to access the channel. The end result of this competition is reduced CR throughput due to collisions among CRs that transmit in the same channel. In this context, the channel sensing order, i.e., the order in which CRs competing for the channels visit those channels, will affect their probability of successful access. This paper presents an investigation into the sensing order dispersion problem, where two or more autonomous CRs attempt to avoid one another by selecting those sensing orders that reduce the likelihood of collisions among one another. We present adaptive strategies that enable the CRs to achieve sensing order dispersion. We evaluate the performance of the presented adaptive strategies in terms of expected time for the autonomous CRs to reach collision-free sensing orders and also in terms of average number of successful transmissions in the network.