Practical unicast and convergecast scheduling schemes for cognitive radio networks

Abstract

Cognitive Radio Networks (CRNs) have paved a road for Secondary Users (SUs) to opportunistically exploit unused spectrum without harming the communications among Primary Users (PUs). In this paper, practical unicast and convergecast schemes, which are overlooked by most of the existing works for CRNs, are proposed. We first construct a cell-based virtual backbone for CRNs. Then prove that SUs have positive probabilities to access the spectrum and the expected one hop delay is bounded by a constant, if the density of PUs is finite. According to this fact, we proposed a three-step unicast scheme and a two-phase convergecast scheme. We demonstrate that the induced delay from our proposed Unicast Scheduling (US) algorithm scales linearly with the transmission distance between the source and the destination. Furthermore, the expected delay of the proposed Convergecast Scheduling (CS) algorithm is proven to be upper bounded by $O(\log n + \sqrt{n/\log n})$. To the best of our knowledge, this is the first study of convergecast in CRNs. Finally, the performance of the proposed algorithms is validated through simulations.