Incorporating Small-World Characteristics into LPWAN: A Comparative Analysis

Low power wide area networks (LPWANs) have gained significant interest due to their cost-effective connectivity solutions for low-power devices spread across extensive geographic regions. They play a crucial role in advancing the Internet of Things (IoT) by enhancing or even surpassing the performance of traditional cellular and short-range wireless technologies. Nevertheless, the challenges associated with a multi-hop LPWAN include increased data latency, inefficient bandwidth utilization, and poor network performance. To address these challenges, a recent breakthrough concept in small-world networks called “small-world characteristics (SWC)” has emerged. This concept achieves a low Average Path Length (APL) and a high Average Clustering Coefficient (ACC) by introducing new long-range links or rewiring existing ones within the network which in turn enhances the network performance. In this study, the focus is on finding the best possible way of incorporating SWC into LPWAN. These methods include different node centrality measures such as degree, betweenness, closeness centrality, and the reinforcement learning technique. Finally, from the developed SW-LPWANs, it is observed that the RL-based technique is showing better SWC in terms of APL and ACC.