Joint Channel Selection and Opportunistic Forwarding in Multi-hop Cognitive Radio Networks

The performance of opportunistic forwarding in the multi-hop cognitive radio networks (CRN) is dependent on the efficient detection of spectrum opportunities and the diversity of relay links. To find appropriate spectrum opportunities, effective spectrum sensing is essential in a CRN, especially under highly dynamic channel conditions. In this paper, a novel channel metric is proposed to determine the sensing sequence at the node based on its history channel observations. This metric is then applied in the design of node residence strategy for relay nodes over the channels. Each node distributively calculates the metric and determines the sensing and residence channel to exploit better spectrum opportunities for packet forwarding. By allowing relay nodes to aggregate in the channel with better transmission opportunity, a larger multi-user diversity gain can be exploited and a lower access delay can be achieved. Simulation results demonstrate that the proposed approach well adapts to the channel dynamics and outperforms other solutions, especially in a CRN with low node density.