Comparison of Channel State Acquisition Schemes in Cognitive Radio Environment

Abstract

We compare the capacity of two most popular methods for acquiring the state of the spectrum in cognitive radio technology: geolocation-database-based schemes and spectrum-sensing-based schemes. For the comparison, we use a new Hidden Markov Chain based channel model, because recent measurements show that a conventional two-state model, which has been widely used, is not appropriate for modeling the behavior of channels in the cognitive environment. We also consider more generic cognitive environments in which each wireless system has its own licensed bands and uses unlicensed bands in addition to its licensed bands to increase its capacity. This type of wireless systems can comprise conventional cognitive radio systems by letting the number of licensed bands be zero. Moreover, we have derived the optimal number of unlicensed bands to be used for maximizing the capacity of wireless system by taking into account interference from neighboring wireless systems. Through simulations, we compare the capacity of wireless systems with two channel state acquisition schemes and show the counterbalancing relation between two schemes, which has never been investigated in previous works. To the best of our knowledge, this is the first work to compare the capacity of these two schemes for acquiring the state of the spectrum.