An effective method of sensor data transmissions in a water-quality monitoring system

In this paper we propose a data selection method based on clustering to reduce the amount of transmissions in a water-quality monitoring system. Recently, some studies are focusing on transmission of the core data collected from the sink nodes of sensor networks. K-means clustering may be a good method for this purpose. Our suggested method, based on K-means clustering, has the outstanding features of selecting unusual data and choosing representative properties of usual data. These features make it possible to reduce the amount of transmissions by removing unnecessary duplication of data. From the experimental results, we found that our method outperformed the related research.