Geographical information based clustering algorithm to equalize cluster lifetime throughout wireless sensor networks

Abstract

Wireless sensor networks (WSNs) usually have a high node density, which makes it unnecessary to require all nodes to be continually active. Scheduling different nodes to be awake at different times saves energy for WSNs. However, traditional scheduling algorithms do not consider the directional data traffic towards the sink and maintaining coverage of the entire network, making them fail to optimize the energy efficiency and the network performance. This paper proposes a clustering algorithm, which combines scheduling techniques, to split the network into clusters based on the geographical information of nodes and the directional data traffic. The clusters are organized in such a way that all clusters have similar lifetimes. Only one node in each cluster is scheduled to be active at any give time to monitor that cluster. Simulation results show that this algorithm not only improves energy efficiency for WSNs, but also improves the performance of delivering data to the sink.