E-BACH: Entropy-Based Clustering Hierarchy for Wireless Sensor Networks

Environmental wireless sensor networks are usually composed of a large group of cooperating wireless sensor nodes spread over an area of interest. They often operate in remote locations and thus must be designed for energy-efficiency and reliability. At the same time, their deployment and operational characteristics must correspond to the spatial distribution of the phenomena of interest. There are a number of approaches that attempt to resolve this trade-off between energy consumption and quality of collected data. Most of them concentrate directly on energy efficiency and communication aspects of network operation. In this contribution, we propose a new method that uses data quality as the primary goal of network optimization. Based on the concept of data entropy, it develops a hierarchical heterogeneous network where individual nodes sample the measured quantities according to the potential information gain. This can be used to save energy by lowering sampling rates of nodes at location with low variability of monitored variables or high correlation with other nodes.