Abstract
Wireless Sensor Network (WSN) is a network of portable and lightweight sensors used to monitor a specific field and report the data they detect wirelessly to a sink node responsible for the analysis and decision making. WSNs have limited power as well as resources. More advanced sensors known as "multimodal sensors" can report more than one feature, which requires even more efficient utilization of the power. Clustering lessens the amount of power lost in WSN. Many clustering algorithms have been proposed for WSNs. However, up to our knowledge, this is the first work that considers multimodal WSNs. In this paper, we propose new techniques for efficient clustering in Multimodal WSN. Through an extensive set of experiments, our proposed algorithms applied to Fuzzy C-Means and K-Means, which are not designed for WSN, have showed an outperformance over LEACH-C, which is a clustering algorithm designed especially for WSN.