POT

Abstract

In this paper, we discuss the top-k monitoring over sensor networks. Since sensor readings are usually correlated with location, top-k nodes are clustered at some areas. Motivated by such a characteristic, we propose a novel tree structure named partial ordered tree(POT) to efficiently maintain clusters of the highest readings. By using POTs, only candidate nodes which might be included in top-k result are evaluated for query processing. Through simulation experiments, we evaluate the performance of the POT method in comparison with conventional methods.