WHAT: A novel routing metric for multi-hop cognitive wireless networks

In this paper, we study how to select a high performance end-to-end path in multi-hop cognitive wireless networks. Due to the time-varying spectrums and undetermined activity of primary users, path selection in cognitive wireless networks is much harder than that of traditional wireless networks. To attack this problem, we propose a novel routing metric, called WHAT (Weighted Hop, spectrum-Awareness, and sTability), which takes not only opportunistic spectrum access into account, but also path stability. WHAT synthesizes channel switching frequency, usage of licensed channels, and length of path together to evaluate the overall quality of an end-to-end path. In addition, different from some existing routing metrics, WHAT has two key properties of routing metric: isotonicity and monotonicity. The theoretical analysis and simulation results show that WHAT can improve the network performance (TCP throughput) profoundly and ensure routing protocols operate correctly.