On energy-aware dynamic clustering for hierarchical sensor networks

Abstract

This paper proposes an energy-efficient nonlinear programming based dynamic clustering protocol (NLP-DC) unique to sensor networks to reduce the consumption of energy of cluster heads and to prolong the sensor network lifetime. NLP-DC must cover the entire network, which is another basic functionality of topology control. To achieve these goals, NLP-DC dynamically regulates the radius of each cluster for the purpose of minimizing energy consumption of cluster heads while the entire sensor network field is still being covered by each cluster. We verify both energy-efficiency and guarantee of perfect coverage. Through simulation results, we show that NLP-DC achieves the desired properties. © IFIP International Federation for Information Processing 2005.