A Sensor Data Collection Method Based on the Constraint of Collected Number of Data and Granularity of Characteristics on an Overlay Network

In this paper, we propose a novel sensor data collection method on an overlay network that forwards messages to the nodes that are suitable for reconstructing a sensor data distribution. When there is an enormous number of sensors, it is redundant to collect all sensor data from a target area since geographically close sensor data can be interpolated. Some methods treat reconstructing a contour lines map by using sensor data collected on overlay networks. However, there is a trade-off between the accuracy and the number of collected sensor data. Our approach is to treat the granularity of the characteristic points in the reconstructed contour lines map in order to satisfy the various requirements of different users. Our proposal method extends the hierarchical Delaunay overlay network (HDOV) and collects a limited number of sensor data. The collection request is forwarded according to the feature amount that corresponds to the characteristics for each layer of the HDOV. Our simulation results showed that our proposal could reconstruct contour lines maps while satisfying the requested granularity of the characteristic points within the number of collected sensor data.