Energy-efficient data collection in mobile sink-based Wireless Sensor Networks using the Hierarchical Clustering Method

Wireless Sensor Networks (WSNs) with mobile sink (MS) were proven to provide extended network lifetime and better data delivery services. This can be achieved by minimizing routing costs and avoiding the development of any hot-spot zones in the network. However, selecting MS visiting points and path determination in WSNs is a challenging task. In this work, We develop an efficient data gathering strategy called EFDC (Energy-Efficient Data Collection) using MS. We determine the optimal cluster heads or visiting points of MS using unsupervised learning-based Hierarchical Agglomerative Clustering, and by using the optimal number of clusters, we can reduce multi-hop communication in a cluster and energy consumption. Also, we propose an efficient trajectory plan among the CHs for the MS using Traveling Salesman Problem with Neighborhoods (TSPN), with minimal computational overhead.