Combining temporal and spatial data suppression for accuracy and efficiency

Reducing the number of data packets transmitted in a sensor network is an effective way of saving energy. Data suppression techniques can reduce data transmissions without losing acceptable data quality. At the data layer, suppression techniques can be classified in two ways, based on spatial data correlation and on temporal data correlation. When employing suppression techniques, a data source node only forwards newly collected data to an aggregating node when the receiver cannot accurately infer the new data values. Inference is based on temporal or spatial data correlations in sensor nodes. In this paper, a novel approach is presented for suppressing data transmissions using both temporal and spatial sensing data correlations. This method offers better performance compared to independent spatial and temporal suppression techniques. Using data from a long-running environmental sensing experiment, we demonstrate how our data gathering approach achieves significant performance gains in terms of both energy conservation and data quality.