Scalability Analysis of the UDCOPA Protocol in Large and Massive IoT Environments

Clustering has a very positive impact on many optimization problems on the Internet of Things and Wireless Sensor Networks. Energy efficiency solutions based on clustering have proved their efficiencies by increasing the lifetime of the networks and the high availability of services provided by applications based on this type of networks. Unequal clustering represents an improvement on equal clustering in terms of flexibility, as it does not impose a predefined radius for clusters’ formation by elected Cluster Heads (CHs). Instead, CHs can dynamically adjust the size of their clusters or the radius of their candidature or election according to various factors and criteria, such as energy constraints. As a result, this type of clustering optimizes energy consumption, balances the load between CHs and improves scalability. Unequal-DCOPA (UDCOPA) is an unequal clustering protocol, which enhances the DCOPA protocol (A Distributed Clustering Based on Objects Performances Aggregation for Hierarchical Communications in IoT Applications), that allows CHs to optimize their energy and send a message announcing their solicitation over an Adaptive Radius of Clustering (ARC) that is adjusted according to its local parameters. In this paper, we explore the geographical and quantitative scalability of this protocol as well as the load balancing of clusters and CHs. The results show that UDCOPA is a scalable protocol that maintains its energy and lifetime properties even in geographically very large areas and in massive environments.