On the impact of primary traffic correlation in TV White Space

In TV White Space, a secondary user must access periodically to a geolocated database to acquire the spectrum availability information. Furthermore, it can access on-demand to the database to update such an information. The more frequent are the on-demand accesses, the higher are the communication opportunities available to the secondary user but the higher is the induced overhead. Hence, in this manuscript, the on-demand access is investigated to a-priori determine whenever it is advantageous to perform it by accounting for the correlation exhibited by primary traffic patterns. To this aim, first the on-demand access is modeled through the general notions of reward and cost. Then, it is proved that the on-demand access maximizing the total reward available to the secondary user is a Markov Decision Process. Stemming from these results, a computational-efficient algorithm is designed. Finally, the theoretical analysis is validated through numerical simulations.