Constructing Budget Connected Dominating Sets in Large-Scale IoT Network Environments

Abstract

Several critical operations depend on various types of information dissemination in the Internet of Things (IoT) networks. To avoid costly flooding-based solutions in these typically large-scale network environments, it is frequently sufficient enough to involve only a proper subset of the nodes (or backbone network), ensuring that every other node will be at most d hops away from a node belonging to the specific subset, i.e., a connected dominating set (CDS). In many cases, the outputted CDS size is upper bounded by a specific physical requirement or constraints relating to the limited available processing and memory capacities of the IoT nodes. Thus, a budget parameter k is frequently introduced as an upper bound for the number of nodes comprising the CDS, leading to the creation of a Budget Connected Dominating Set (BCDS). In this work, a novel distributed algorithm for constructing a BCDS is proposed based on information available d hops away from the network nodes, assuming an upper budget k limit regarding its final size. An additional upper bound parameter relating to the number of sent messages for each node, i.e., the stop, is also for the first time utilized. It is shown using simulation results that when stop is employed, a BCDS with increased coverage (i.e., the fraction of nodes at most d hops away from the BCDS nodes) is constructed under the proposed algorithm.