An Overhead Free Clustering Algorithm for Wireless Sensor Networks

An overhead-free, fully distributed clustering algorithm is proposed to decompose wireless sensor networks, where nodes are initialized with either equivalent or different energy capacities, into a two-tier clustered hierarchical structure. Energy-rich nodes are assured to act as cluster heads (CH), and CHs are dispersed evenly over the network. In the new algorithm, a converting function, a multiplicatively increasing CH selection probability, and two backoff strategies are interwoven over three phases during the CH selection and placement. Via simulations, the performance of the proposed algorithm has been demonstrated considering representative network scenarios. The results show that our algorithm outperforms some existing clustering methods in extending the system lifetime and enlarging the network data capacity.