An Efficient Cluster-based Self-organization Algorithm for Wireless Sensor Networks

International audienceThe paper proposes an Efficient Cluster-based Self-organisation Algorithm (ECSA) for partitioning Wireless Sensor Networks (WSNs) into clusters, thus giving at the network a hierarchical organisation. Each sensor uses its weight based on its k-density and its residual energy to elect a cluster-head in its 2-hop neighbourhood. ECSA operates without a centralised controller, and does not require that the location of sensors be known. Firstly, we estimate the number of clusters formed with ECSA according to transmission range, then we evaluate the amount of packets sent to the sink per energy dissipation by using the same model presented by Heinzelman et al. (2002), and the number of nodes alive per number of packets received at the sink. Simulation results illustrate that ECSA can evenly distribute energy consumption of sensors and consequently maximise network lifetime when compared to LEACH and LEACH-C