Distributed WSN Data Stream Mining Based on Fuzzy Clustering

This paper proposes a distributed wireless sensor network (WSN) data stream clustering algorithm to minimize sensor nodes energy consumption and consequently extend the network lifetime. The paper follows the strategy of trading-off communication for computation through distributed clustering and successive transmission of local clusters. We present an energy efficient algorithm we developed, Subtractive Fuzzy Cluster Means (SUBFCM), and analyze its energy efficiency as well as clustering performance in comparison with state-of-the-art standard data clustering algorithms such as Fuzzy C-means and K-means algorithms. Simulations show that SUBFCM can achieve WSN data stream clustering with significantly less energy than that required by Fuzzy C-means and K-means algorithms.