A Predictive Opportunistic Access Scheme for Cognitive Radios

We propose a novel opportunistic access scheme for cognitive radios that considers the channel gain as well as the predicted idle channel probability. In contrast to previous works where a cognitive user vacates a channel only when that channel becomes busy, our scheme requires the cognitive user to switch to the channel with the next highest idle probability if the current channel´s gain is below a certain threshold. We derive the threshold values that maximize the long term throughput for various transition probabilities and compare our throughput results to those obtained using a non-channel adaptive scheme. We also provide simulation results that validate our analytical results.