A Dynamic Data Aggregation Scheme for Wireless Sensor Networks

Wireless sensor networks are formed of tiny, energy-constrained sensor nodes that could be mobile and may be deployed in unfamiliar environments in large numbers. Considering these unique characteristics, our network architecture is modelled around a data-centric approach that allows us to make use of in-network processing and data aggregation which in turn helps to maximize network lifetime. This paper suggests methods to improve network efficiency by combining Directed Diffusion [2] with clustering and by introducing a more elaborate data aggregation scheme. Our data aggregation scheme allows nodes to process data collected from sensors and subsequently aggregate the data even in completely unfamiliar environments by including entire query definitions within interest messages. We also describe certain novel design features such as interest transformation, layered data aggregation and dynamic data aggregation points all of which would improve overall system performance.