A New Architecture for Cognitive Radio Networks Platform

Cognitive radio networks is recognized as the technology to solve the problem of spectrum shortage and promote the efficient use of the spectrum. However, the research of CR networks technologies only stays in the theoretical stage recently. Therefore, developing a CR networks platform becomes an urgent problem to solve. Several CR platforms have being deployed by the research community. However, most platforms provide cognition either in physical layer or network layer. Few platform addresses all layers cognition for wireless network. This paper proposes a new architecture which is aimed at stimulating cognition process in all layers for wireless networks. We firstly review the current main CR platforms. Then, a new CR networks platform is proposed. We clearly describe the components of the new platform and give work flow of the new platform. Finally, we use the platform to simulate and verify two spectrum assignment algorithms, and give performance comparison. Simulation results indicate that the greedy algorithm is better than fair algorithm in spectrum utilization and the greedy algorithm outperforms the greedy algorithm in fairness.