A Clustering Optimization for Energy Efficiency in Wireless Sensor Network Using K-Means Algorithm

Abstract

The limitation of sensors energy make energy efficiency still a priority issue in Wireless Sensor Network (WSN) technology. One effort that can be done to overcome this problem is to design the right data transmission path (or better known as routing). Low Energy Adaptive Clustering Hierarchy (LEACH) is one of the most widely used cluster-based routing protocols because it is considered capable of minimizing the amount of energy consumption through the formation of clusters or groups of nodes. Unfortunately, this protocol will experience a significant decrease in energy as the amount of data transmission increases. This is partly due to the clustering process which is carried out randomly and causes an imbalance in the distribution of the number of nodes between clusters. This study proposed a method to optimize the clustering process in the LEACH protocol by integrating the K-Means algorithm, which is called LEACH-KMe. A simulation was conducted to determine the effectiveness of the proposed method by considering 4 main parameters, namely total energy consumption, number of alive nodes, number of dead nodes, and residual energy. The test results proved that the LEACH-KMe protocol provides better performance than the conventional LEACH protocol (more even distribution of nodes, less total energy consumption and number of dead nodes, as well as a larger number of alive nodes and residual energy).