Low Message Overhead Clustering Algorithm for Wireless Sensor Networks

Abstract

A Wireless Sensor Network (WSN) is defined by a set of sensor nodes and a Base Station (BS). Typically, sensor nodes are distributed over a geographical area, sense some information, process it and send it to the BS via a wireless link. Since sensor nodes are small, may be distributed in large scale or in a dangerous area, their battery are small and may not be recharged or replaced. So the network lifetime is prolonged when the battery energy is used wisely. The function of the BS is to receive information sensed by the nodes, therefore it's a sink. One way to send information is to cluster the WSN, dividing the sensor nodes in two exclusive sets: Cluster Heads (CH) or Cluster Members (CM). Therefore, the clustered WSN presents two types of communication: one from CM to CH, inter-cluster communication, and other from CH to BS, intra-cluster communication. Once a WSN is clustered, energy consumption and coverage are improved. The first step in clustering a WSN is a CH election algorithm. We propose a distributed clustering algorithm for WSN named Low Message Overhead Clustering Algorithm (LMOCA) that has the goal to spend as minimum energy as possible on the clustering process. Beyond that it is very intuitive and easy to implement.