A Framework to Support Continuous Range Queries over Multi-Attribute Trajectories

Emerging applications over spatio-temporal trajectories require representing the data from diverse aspects. We study multi-attribute each of which consists a sequence time-stamped locations and set attributes characterizing investigate continuous range queries trajectories. Such query returns whose contain expected values are always within distance threshold to trajectory during entire overlapping time period. To efficiently answer query, an optimal method partitioning is proposed index structure developed support combined search using both parameters attribute values. Query algorithms auxiliary structures developed, accompanied with optimization strategies thorough theoretical analysis. Using real synthetic datasets, we carry out comprehensive experiments in prototype database system evaluate efficiency scalability our designs. The experimental results show that approach outperforms six alternative approaches by factor 5-50x on large datasets.