Mobility Adaptive Clustering Algorithm for Wireless Sensor Networks with Mobile Nodes

In this study, the authors propose a mobility adaptive clustering algorithm for wireless sensor networks with mobile nodes. In the proposed algorithm, a sensor node selects itself as a cluster head based on a Single Point Predictor for the combined criterion prediction. Here, as a combined criterion predicting proposed comprising parameters as connectivity, coverage, mobility and residual energy consumption of all the nodes (Distributed Clustering Algorithm). A non-cluster-head nodes declare themselves as members of the cluster based on a new value, that value is related to each cluster head, which is used to indicate its suitability for connecting this cluster head or other (Mobility-Based Clustering protocol). We found in the Distributed Clustering Algorithm the Mechanism to elect the head of the cluster and in the Mobility-Based Clustering protocol the mechanism for engagement a non-cluster head nodes to a specific cluster. Thus, the combination of them together provides a significant increase in the network life-time, availability, stability period of wireless sensor networks and reduction of packet loss in the cluster formed by a simple point predictor for combined criterion prediction and stability value.