Regresar

Cluster-Head Selection for Energy-Harvesting IoT Devices in Multi-tier 5G Cellular Networks

Abstract:

Fifth-generation cellular networks promise to interconnect a wide variety of wireless devices, such as pico cells, femto cells and Internet of Thing devices. In this vision, multiple tiers appear as one of the main solutions to overcome these challenges providing an increase in the spectrum efficiency through the spectrum reuse. In this work, we investigate an approach to increase the lifetime of Internet of Thing devices, equipped with energy-harvesting capabilities, in a heterogeneous network considering a macro-tier and an underlay-tier. The proposed solution considers the k-means algorithm to group the IoT devices and a cluster head selection algorithm to balance the energy consumed in the network based on the residual energy and distance to the regional manager node. To improve the knowledge of the amount of energy harvested, we propose a solar intensity-pbkp_rediction approach based on a multilayer perceptron algorithm from which the results are used to increase the performance in the cluster head selection scheme. Numerical results show that the proposed approach achieves significant improvement over the baseline Low Energy Adaptative Clustering Hierarchy scheme in increasing the residual energy and the number of IoT devices alive in the network.