-
Essay / A new clustering scheme for spectrum sharing in...
Spectrum scarcity can be addressed using the cognitive radio networks (CRN) paradigm. In CRNs, secondary users (SUs) without spectrum licenses can benefit from spatio-temporally vacated spectrum opportunities left by primary users (PUs) if these SUs do not cause any harmful interference to the PUs. Due to the ease of deployment, ad hoc CRNs are expected to attract future applications of secondary spectrum use [1]. However, the new challenges of ad hoc CRNs must be addressed first to visualize the new revolution of this type of networks. Variation in spectrum availability over time and locations creates a unique trait of CRNs. This key feature poses great challenges from different aspects in the basic design of ad hoc CRNs. One of the major challenges is how SUs can exchange their control signals to coordinate their spectrum sensing, spectrum allocation and access, and traffic routing. The intuitive method is to have a common control channel (CCC), which can be an in-band or out-of-band channel. In the literature, there are different approaches to choosing CCC, and readers can refer to [1] for more details. However, the spectrum heterogeneity observed by SUs in multi-hop ad hoc CRNs makes the use of a CCC an embarrassing challenge. Network clustering can be used to overcome spectrum heterogeneity by grouping nearby users who may experience the same activities from the same set of Pus; Additionally, clustering can help reduce the signaling overhead needed to operate the network and maintain its connectivity. Almost all existing clustering approaches in CRNs aim to avoid using a global CCC and instead rely on intra-band CCCs to coordinate their processes and form clusters...... middle of paper.. ....stems expected to cover a larger area of ​​communications than SUs, so each group of nearby SUs should be almost under the coverage of the same set of PUs. Nearby SUs are clustered with 1 hop of a central user based on the geographic locations of the users and their temporal occurrence orders, such that each cluster covers a geographic portion of the entire communication region. SUs in each cluster cooperate to sense the spectrum in their neighborhood and then access available channels using a MAC framework. As shown in Figure 1, each cluster consists of a Cluster Head (CH) user who initiates the cluster, Cluster Member (CM) users, and Cluster Gateway (CG) users who join the cluster with its neighboring clusters. . We assume that SUs rarely move, and when they do, they move slowly, so the network topology is dynamic with stable status..