Experiences with Implementing a Distributed and Self-Organizing Scheduling Algorithm for Energy-Efficient Data Gathering on a Real-Life Sensor Network Platform

We report our experiences with implementing a distributed and self-organizing scheduling algorithm designed for energy-efficient data gathering on a 25-node multihop wireless sensor network (WSN). The algorithm takes advantage of spatial correlations that exist in readings of adjacent sensor nodes and utilizes cross-layer information from the underlying MAC layer to minimize message transmissions. We describe how we modify our experiments in order to meet the assumptions made in the earlier theoretical analysis of the algorithm. The implementation results which are virtually identical to the preliminary simulation results, show that the algorithm achieves up to 80% energy savings when compared to conventional raw data collection.