Range Queries on Multi-Attribute Trajectories

jianqiu Xu, Hua Lu

Abstract:

Motivated by the trend of providing comprehensive knowledge about
trajectory data, we study multi-attribute trajectories each of which contains a
sequence of time-stamped locations and a set of characteristic attributes. This
enriches the data representation by providing a comprehensive description of moving
objects and thus enables new types of queries onmoving object trajectories. In this
paper, we consider answering range queries that return trajectories (i) containing
particular attribute values and (ii) passing a certain area during the query time.We
integrate standard trajectories and attributes into one unified framework and propose
an index structure as well as the query algorithm. The structure is general and flexible
in terms of handling bothmulti-attribute trajectories and standard trajectories,
answering a range of queries and supporting update-intensive applications. The
evaluation is conducted in a prototype database system and experimental results
demonstrate that our method outperforms alternative methods by a factor of 3-10
on a data set of one million real trajectories and synthetic attribute values.