Optimal Channel Selection in Multi-User Multi-Channel Secondary Networks

Maximizing a secondary network's throughput, that relies on spectrum sensing for band acquisition, is the focus of this work. Users of a secondary network may attempt to acquire the best bands based on prior information. Without any cooperation among users during or after sensing, multiple users may attempt to sense/acquire the same band leading to a loss in throughput due to contentions. We propose a continuous channel selection strategy and show that it maximally utilizes the opportunity in the physical layer which was left untapped by the existing works. The throughput performance of the proposed continuous channel selections in both low and high opportunity primary activity profiles outperforms the discrete selection based design. We also propose a low complexity algorithm to achieve optimal channel selection among users. Numerical results show that system throughput with continuous channel selections is better than with discrete channel selections. In the worst case interference scenario, continuous selections perform as good as discrete selections.