An EH-WSN Clustering Algorithm Based on Energy Prediction

Abstract

The energy-constrained problem of wireless sensor networks is harmful to its application in many fields like fire monitoring, smog monitoring and smart city. Traditional energy-saving methods such as node dormancy, mobile SINK, etc. cannot fundamentally solve the energy-constrained problem, and sometimes even lead to negative consequences such as smaller network coverage or reduced throughput. The use of surrounding renewable energy sources is a great solution to solve the problem. However, it will create a new problem because of instability of renewable energy sources. So the effective use of renewable energy to supplement sensor energy is inseparable from accurate energy prediction algorithms. This paper proposes an adaptive filter-based prediction algorithm. And based on the prediction algorithm, a node clustering method is proposed, which performs well in the energy harvesting wireless sensor network, solving the problem of maladjustment of DEEC in EH-WSN. Experiments show that compared with the DEEC algorithm, the live nodes increase 17.1% in the 230th round.