Incremental In-Network RNN Search in Wireless Sensor Networks

With the rapid advances of wireless communication and sensor technologies, spatial queries on moving objects are increasingly important in many sensor applications. One of the most frequently used spatial queries is the reverse nearest neighbor (RNN) query that returns the objects whose nearest neighbor is the query object. RNN answers are particularly essential to the users who are under emergency circumstances. In this paper, we propose an incremental RNN search (IRS) method to answer RNN queries for moving objects in sensor networks. IRS employs a filter-verification framework to achieve the energy-efficient query processing by filtering out most objects irrelevant to the query results. To support the execution of IRS, we also design a distributed object location management scheme to reduce the amount of communication in managing moving objects. Our experimental results reveal that IRS is indeed quite promising, as it requires only a moderate amount of communications for managing moving objects as well as for processing RNN queries.