NobleBlocks

Research indexJournal article

МЕТОД АДАПТИВНОЇ КЛАСТЕРИЗАЦІЇ МОБІЛЬНИХ ВУЗЛІВ ТУМАННОГО ШАРУ ІНТЕРНЕТУ РЕЧЕЙ

Kovalenko, Kuchuk

Системи управління навігації та зв’язку Збірник наукових праць18 Sep 2026Open access

NobleIDark:/48914/w1/20261105/807A74F7DOI10.26906/sunz.2026.3.106

Abstract

Relevance. The primary criterion for clustering fog devices is the proximity of the fog nodes belonging to a cluster. However, when mobile nodes are present, the cluster structure may be disrupted if a mobile node moves beyond the cluster boundaries. This may result in unexpected delays in IoT data processing. Therefore, for such systems, an additional criterion related to the stability of the cluster structure should be considered. The subject of the study is clustering methods for high-density foglayer Internet of Things networks. The aim of the study is to improve the stability of mobile components of the fog layer by modifying the clustering approach through the development of a method for forming a stable mobile cluster. The objectives of the study are to develop an approach to the architecture of the mobile component, as well as a model and a method for ensuring the stability of a mobile cluster in a high-density fog-layer Internet of Things network. Research results. To organize the interaction of mobile devices, an architecture of the mobile component of the fog layer has been developed, taking into account its decentralized structure and the specific operational characteristics of mobile Internet of Things devices. The proposed architecture consists of four levels and provides for the formation of separate mobile clusters of fog devices at the lower level. To investigate the formation of mobile clusters, a corresponding model has been developed that accounts for the random nature of the movement of Internet of Things devices. The spatial distribution of mobile components is modeled using the Thomas point process. Based on the proposed model, the FOREL clustering algorithm has been modified. In the modified algorithm, the position of the cluster center is determined using weighting coefficients that characterize the stability of its components. Conclusions. Simulation results demonstrated increased stability of the formed mobile clusters in environments with high and ultra-high device density. It was found that an increase in the mean relative deviation of the actual device trajectories from their planned trajectories leads to an increase in the stability of the cluster structure. This behavior is explained by the displacement of the cluster center toward its least stable components, which enables the cluster structure to adapt to changes in the spatial distribution of mobile devices. The proposed method can be used for clustering mobile components of the fog layer of the Internet of Things when the mean deviation of device movement from the specified trajectories does not exceed 20% of the cluster radius.

Discussion

Citations

More

Suggest a correction or report