GIT-CERCS-11-08
Myungcheol Doo, Ling Liu,
Spatial Alarm Processing and Algorithms
Spatial alarms are fundamental capability for location based advertisements and location based reminders. One of the most challenging problems in scaling spatial alarm processing is to compute alarm free regions (AFR) such that mobile objects traveling within an AFR can safely hibernate the alarm evaluation process until approaching the nearest alarm of interest. In this paper we argue that maintaining an index of both spatial alarms and empty regions (AFR) in the context of spatial alarm processing) is critical for scalable processing of spatial alarms. Unfortunately, conventional spatial indexing methods, such as R-tree family, k-d tree, Quadtree, and Grid, are not well suited to index empty regions. We present Mondrian Tree - a region partitioning tree for indexing both spatial alarms and alarm free regions. We first introduce the Mondrian tree indexing algorithms, including index construction, search, and maintenance. Then we describe a suite of Mondrian tree optimizations to further enhance the performance of spatial alarm processing. Our experimental evaluation shows that the Mondrian tree index outperforms traditional index methods, such as R-tree, Grid, Quadtree, and k-d tree, for spatial alarm processing.