Cluster-based efficient information dissemination in dynamic networks

Abstract

Dynamic network is an abstraction of networks with frequent topology changes arising from node mobility or other reasons. In this article, we first propose a dynamic network model, named (T, L)-HiNet, to extend existing dynamic network models with clusters. (T, L)-HiNet includes several properties defining the stability of cluster hierarchy in a dynamic network, including cluster head set, cluster member set and the connections among them. Based on (T, L)-HiNet, we design several hierarchical information dissemination algorithms for different scenarios of dynamics. Furthermore, we extend the (T, L)-HiNet model and corresponding algorithms in two directions, that is, stability of cluster head set and cluster size. The correctness of our algorithms is proved rigorously, and their performance is evaluated via both numerical analysis and simulations. The results show that compared with the algorithm recently proposed by Kuhn et al., our design can significantly reduce communication cost and also time cost.