Heuristic Spectrum Assignment Algorithm in Distributed Cognitive Networks

Cognitive radio is an exciting emerging technology that has the potential of dealing with the urgent requirement and scarcity of the radio spectrum. Although having multiple radio interfaces and available spectrum bands can generally increase the effective throughput, a problem arises as to what the best strategy to dynamically assign available bands to secondary users for maximizing throughput by minimizing the interference, and what the best scheme to allocate the spectrum holes to unlicensed users to maximize the fairness. This paper presents a distributed and heuristic spectrum assignment algorithm for multi-radio wireless cognitive networks in a cognitive network environment. The proposed algorithm (Fairness Bargaining with Maximum throughput, FBMT) considers the problems including system throughput and the fairness. Extensive simulation studies in 802.11 based multi-radio cognitive networks have been performed. The results indicate that the proposed algorithm can facilitate a large increase in network throughput and acquire a good fairness performance in comparison with a common spectrum assignment mechanism that is used as a benchmark in the literature.