Spectrum-aware Distributed Channel Assignment for Cognitive Radio Wireless Mesh Networks

In Cognitive Radio Networks, the application throughput is not only affected by primary user activity but also by numerous environment factors such as interference. Therefore, channel assignment for cognitive radio networks should not only consider channel idle time but also an error rate perceived on the channel. The spectrum-aware channel assignment is vital to efficiently utilize the network resources. In this paper, we propose Spectrum-aware Channel Assignment (SaCA) algorithm for multi-radio, multi-channel cognitive radio networks. We have simulated our proposed algorithm in OMNeT++, an open source discrete event simulator, and compare its performance with the spectrum-unaware channel assignment (SuCA) algorithm. The performance of channel assignment is evaluated for packet delivery ratio and number of channel switches by varying the number of primary users, number of channels and primary user activity ratio. The performance of SaCA is better for large number of channels, primary users and higher primary user activity ratio in the network. In comparison with SaCA, average packet delivery ratio more sharply decreases with increase in number of primary users for SuCA.