Entropy Analysis of Hierarchical Routing in Heterogeneous Wireless Sensor Network

Sparing energy during data transfer in Wireless Sensor Networks (WSN) requires the main communication facility of the Internet of Things (IoT) services. The hierarchical approach of cluster preparation is considered a de facto mechanism in the WSN systems. The creation of the clusters and election of the cluster head (CH) remains one of the most intensively analyzed mechanisms because the interference of the sensor nodes strongly affects the system performance. The entropy measures the uncertainty level of a system. Disjunctive subsystems may or may not be correlated. The more the subsystems are independent, the more approaches from above the sum of subsystems’ entropy to the entropy of the union system. The spatiotemporal distribution of the CH has a big impact on the consumed energy by the WSN radio interfaces and the overall life cycle of the sensor communication system. A typical instance of the two-level hierarchical routing mechanisms is the Low-Energy Adaptive Clustering Hierarchy (LEACH) considering just the physical distance between the potential members of the clusters. We proposed a Cost-Balanced LEACH (CB-LEACH) with a moving Sink node (SN) in the previous study. Both the distance and the remaining energy of the nodes were used in the CH election step. We proved previously that CB-LEACH significantly outperformed the classical LEACH when a cost balance metric based on the actual global status of the WSN (mean and standard deviation of the nodes’ energy and the nodes’ relative distance) was applied in the CH election step. In this study, heterogeneity of the analyzed WSN was marked by two groups of sensor nodes spread uniformly randomly in the same physical area, each group of nodes having different amounts of initial energies. Based on the newly proposed metric called Inter-Subsystem Mutual Effect (ISME), we investigated the effect of the heterogeneity in the CB-LEACH system using energy-entropy (EE) and distance-entropy (DE). The relative difference of these entropies indicates the interference of the subsystems.