Reverse Nearest Neighbor (RNN) query is to find the set of objects that are closer to the Q than any other objects in dataset D. Owing to the wide application spectrum, there have been great demands for effective RNN query processing in the circumstance where the sensor nodes are deployed over a wide geographical area and track the location of objects. However, relentless energy and computing resource depletion are inevitable by the maintaining the infrastructures in the existing researches. Motivated by these issues, we propose a novel approach, named the parallel itinerary-based RNN (PIRNN) query processing algorithm which does not rely on any kind of infrastructures. PIRNN disseminates multiple itineraries concurrently and it prunes the search area to increase performance. Furthermore, we extend PIRNN with two optimization heuristics, called Peri-Segment Completion (PSC) and Look Forward (LF) to minimize the area to be searched. In order to evaluate the performance of PIRNN query processing, we compare PIRNN with itinerary-based SAA and TPL. The extensive simulation results show that the PIRNN method outperforms SAA and TPL in terms of network traffic.
- Reverse nearest neighbor
- Spatial query
- Wireless sensor network
ASJC Scopus subject areas
- Electrical and Electronic Engineering