Channel rendezvous is a vital step to form a cognitive radio network (CRN). It is intractable to guarantee rendezvous for secondary users (SUs) within a short finite time in asynchronous, heterogeneous and anonymous CRNs. However, most previous heterogeneous algorithms rely on explicit SUs’ identifiers (IDs) to guide rendezvous, which is not fully distributed. In this paper, we exploit the mathematical construction of sunflower sets to develop a Single-radio Sunflower- Sets-based (SSS) pairwise rendezvous algorithm. We propose an approximation algorithm to construct disjoint sunflower sets. Then SSS leverages the variant permutations of elements in sunflower sets to adjust the order of accessing channels instead of SUs’ IDs, which is more favorable for anonymous SUs in distributed environments. We also propose a Multi-radio Sunflower-Sets-based (MSS) pairwise rendezvous algorithm in order to bring additional rendezvous diversity and accelerate the rendezvous process. Moreover, for the case with more than two SUs, we propose a multi-user collaborative scheme in which SUs cooperatively exchange and update their channel-hopping sequences until rendezvous. We derive the theoretical upper and lower bounds of rendezvous latency of the proposed algorithms. Extensive simulation comparisons with the state-of-the-art blind rendezvous algorithms are conducted incorporating the metrics of maximum and expected time-to-rendezvous. The simulation results show that our algorithms can achieve rendezvous faster than previous works.

This work was published on IEEE Transactions on Vehicular Technology, 2016, pp1-14. titled Fully Distributed Channel-Hopping Algorithms for Rendezvous Setup in Cognitive Multi-Radio Networks