Network lifetime is always a challenging issue in battery-powered networks due to the difficulty of recharging or replacing nodes in some scenarios. Clustering methods are a promising approach to tackle this challenge and prolong lifetime by efficiently distributing tasks among nodes in the cluster. The present study aimed to improve energy consumption in heterogeneous IoT devices using an energy-aware clustering method. In a heterogeneous IoT network, nodes (i.e., battery-powered IoT devices) can have a variety of energy profiles and communication capabilities. Most of the existing clustering algorithms have neglected the heterogeneity of energy capacity among nodes and assumed that they are of the same energy level. In this work, we present HetEng, a Cluster Head (CH) selection process that extended an existing clustering algorithm, named Smart-BEEM. To this end, we proposed a statistical approach that distributes energy consumption among highly energetic nodes in the network topology by constantly changing the CH role between the nodes based on their real energy levels (in joules). Experimental results showed that HetEng resulted in a 6.6% increase of alive nodes and 3% improvement in residual energy among the nodes in comparison with SmartBEEM. Moreover, our method reduced the total number of iterations by 1% on average.
View on arXiv