Data Analysis in Wireless Sensor Networks with Distributed Self Organizing Map
Anita Panwar, Satyasai Jagannath Nanda · 2024
Distributed clustering algorithms are employed in wireless sensor network (WSN) to improve the local data analysis. This process is carried out collaboratively with the help of nearby neighbours without a central controller. In this paper, distributed clustering is performed with Self Organizing Map (SOM). The SOM is a popular unsupervised neural network model that maps input data to a lower-dimensional grid. On this grid map similar input patterns are placed closer to each other. This process helps in discovering patterns and relationships in the data without prior labeling, thus making proposed Distributed Self Organizing Map (DSOM) useful for unknown local data analysis at the WSNs. The proposed algorithm is applied to analyze two real life WSN datasets: Water quality monitoring of Thames river, Weather monitoring dataset of various stations at Canada. Comparative analysis is carried out with Distributed Particle Swarm optimization algorithm and Distributed K-means algorithm. The proposed DSOM has superior performance, as indicated by Silhouette Index and Quantization Error measurements.