Utilizing a Multimodal Wireless Node Features for the Benefit of WSN Lifetime

Wireless Sensor Network (WSN) is considered nowadays as one of the promising research areas due to the wide variety of applications that it can serve. WSNs attracted many of the researchers as well as industry towards devising methods to increase the stability, longevity and functionality that could be delivered by WSN. WSN could be used to detect motion, temperature and humidity, which allow them to be used in e.g. security applications, weather monitoring just to name a few. The network consists of lightweight portable nodes that are deployed in a huge number with a built-in power source normally a battery. Recharging operation to these nodes could be considered impossible due to the large number of deployed nodes and the difficulty to have access to the nodes after deployment. This required efficient usage of the available power resources which directed researcher to devise algorithms and methods by which the power consumption is minimized; one of these methods is clustering. Clustering is a technique that is used in a number of fields like image processing and data analysis. It has been adapted to WSN to serve the specific characteristics of WSN. Several algorithms have been proposed specifically for WSN but the proposed algorithms didn't consider the ability of the node to sense more than one parameter e.g. temperature and humidity. In our work we have considered the idea of using Fuzzy C-Means (FCM), which is clustering algorithms, that allows a node to be assigned to more than one group with different membership percentage. We used FCM with applying our proposed algorithms, which showed a clear outperformance of FCM using our proposed algorithms over Leach-C, which is a clustering algorithm specially designed for WSN.