Aware channel assignment algorithm for cognitive networks

Abstract

Several recent studies have shown that Cognitive Networks (CNs) can effectively address the spectrum shortage problem in wireless networks, which is mainly caused by the increasing number of wireless services and applications operating in unlicensed channels. However, these studies fail to consider the differences between channel characteristics, i.e. they do not consider that channels are on different frequencies. In this paper, we fill this gap by providing a technique that classifies channels based on their operating frequency. We enable each cognitive device to choose the best channel depending on its traffic demand. We focus on the IEEE 802.22 physical layer in order to analyze and classify channels, and we propose the aware channel assignment algorithm for cognitive networks (Aaron). Aaron assigns channels to cognitive devices with the goal of satisfying the capacity demand of the largest number of end-users in order to maximize the throughput. We evaluate Aaron using an ad-hoc event-driven simulator for CNs. In addition we compare it with the Dumb algorithm, where cognitive devices are not able to characterize channels, and with the Upper Bound, where no packet is lost due to the channel assignment algorithm. Simulation studies demonstrate that Aaron performs considerably better than Dumb and very close to the Upper Bound.