Semijoin-Based Spatial Join Processing in Multiple Sensor Networks

This paper presents an energy-efficient spatial join algorithm for multiple sensor networks employing a spatial semijoin strategy. For optimization of the algorithm, the authors propose a GR-tree index and a grid-ID-based spatial approximation method, which are unique to sensor networks. The GR-tree is a distributed spatial index over the sensor nodes, which efficiently prunes away the nodes that will not participate in a spatial join result. The grid-ID-based approximation provides great reduction in communication cost by approximating many spatial objects in simpler forms.